Skip to main content

Full text of "The Founder Of Modern Egypt"

See other formats



OU 218409 

DO — 
73 < 

> m 

73 73 


Gall Mo. , Accession No, / ??^/ 


^. il|e I’d i 

This book should be returned on or before the date last marked below , 



Dupleix and Clive , 
The Nabobs of Madras, 
etc., etc. 


Joint Editor of 
The Cambridge History 
of India 






Professor of the History and Culture of 
the British Dominions in Asia , in 
the University of London 

“jf* n' at fait en Egypte que ce que les Anglais 
ont fail aux Indes ” 




Cambridge University Press! 
Fetter Lane, London 

New Tork 

Bombay , Calcutta , Madras 

Maruzen Company, Ltd 

All rights reserved 



Preface page vii 

Chap. I The Rise of Muhammad ’Ali I 

II The Pillar of the Empire: Arabia and 

the Sudan 39 

III The Pillar of the Empire: the Greek War 68 

IV The Algerian Interlude and the Con- 
quest of Syria 94 

V The Idea of an Arab Empire and the 

Overland Routes 1 25 

. V 

VI The Second Syrian War and the collapse 

of Muhammad ’Ali’s plans 154 

VII Muhammad ’Ali’s Government in Egypt 192 

VIII Muhammad ’Ali’s Government in Crete 

and Syria 242 

Conclusion 259 




The following pages constitute an endeavour to escape 
from the traditional hero of French and villain of 
English writers, and to ascertain by a study of dHginal 
materials what Muhammad ’Ali really did. In many 
respects this task has become much easier of late years 
than ever it was before. A large amount of most im- 
portant matter has been published by the Societe 
Royale de Geographic d’Egypte, under the wise 
patronage of His Majesty King Fouad. Its publications 
include French, English and Italian documents of 
great interest and value. 

Besides these I have examined in detail the reports of 
our own agents preserved among the records of the 
Foreign Office and the India Office. 

By the courtesy of M. Cattaui I have been enabled 
to make use of the yet unpublished reports of the Rus- 
sian consuls-general, especially those of Count Medem, 
the Russian agent during the most critical period of 
Muhammad ’Ali’s career. 

Lastly, by the permission of His Majesty King Fouad, 
I have been enabled to examine valuable series of letters 
and orders issued by Muhammad ’Ali to his principal 

In this connection I must acknowledge the special 
assistance of M. Deny and M. Gelat Bey, who always 
found time to afford me help whenever I needed it in 
the course of my work at Cairo. 



I owe special acknowledgements to M. Georges Douin 
and to Professor L. M. Penson — to the first for the ex- 
tensive use I have made of the volumes which he has 
contributed to the series published by the Societe Royale 
de Geographic d’Egypte, and to the second for reading 
my proofs and favouring me with valuable criticisms. 

H. D. 

School of Oriental Studies , 

Finsbury Circus , London . 

Chapter I 


We are perhaps still disposed to undervalue the force 
and originality of our eighteenth-century ancestors. 
Their formal manners, their decorative dress, their 
heroic couplets, their sentimental novels, their super- 
ficial histories, easily suggest the sophisticated ending 
of an old rather than the beginning of a new world. Yet 
it is hard to over-estimate our modern debt to them. 
They conceived and handed down to us not only 
humanitarian ideas and evolutionary theory, but also 
the application of steam to industry and a most com- 
plete revolution in the art of war — the great points on 
which our modern history and ideas have turned. In 
fact they revolutionised the sources of power. And in 
so doing they begot the ruin of the great oriental 
empires. Power was no longer bestowed by the de- 
scendants of those nomad tribesmen who had burst east, 
west and south from the steppes of Central Asia, carry- 
ing in their train havoc and empire. It now belonged 
to the western races, with their highly disciplined in- 
fantry able without fear to confront any horsemen, with 
their siege guns able to blast a way through walls 
however stout, with their fieldpieces that could scatter 
any concentration of Asiatic cavalry. By the close of 
the period the states of India had already experienced 
the force of the new weapon and were bending before 
it; and in the Nearer East the Turks, who had not so 
long ago pierced the Carpathians and almost taken 
Vienna itself, could no longer hold their own. Their 
frontier guards were being driven in; province after 
province was being wrested from them; their hold on 


Constantinople itself was relaxing. And the growing 
sense of military inferiority was carrying with it a 
multitude of moral consequences. As self-confidence 
vanished, mutual confidence weakened. The sar-’askar 
— the commander of the host — mistrusted his lieu- 
tenants, and was mistrusted by his men. That Might of 
Islam to which such noble mosques had been dedi- 
cated in the past was visibly sinking to the ground. Even 
the despised Christian subjects, who for centuries had 
tilled the fields and paid the infidel poll tax as un- 
murmuringly as the Hindu subjects of Delhi, were 
beginning to raise their heads and whisper of indepen- 
dence, and the pashas of the Sultan, like the nawabs and 
nazims of the Moghul Emperor, now executed those 
orders only that were beneficial to themselves. The 
pashaliqs of Baghdad, of Damascus, and of Cairo were 
hardly more than nominal dependencies. 

The province of Egypt had indeed long been suffering 
from a more than Turkish misrule. Its relations with 
the empire, even in the early days following Salim’s 
conquest, had always been of the loosest. It had been 
entrusted to the tyranny of such of the Mamelukes as 
had survived massacre and had proved their unworthi- 
ness by deserting their own master. Over them in- 
deed was set a pasha appointed from Constantinople, 
often changed, but seldom more than a governor in 
name. The beys, as the Mameluke chiefs were called, 
sought nothing but personal and private ends. Their 
followers, recruited as slaves from Circassia and Georgia, 
were sedulously trained as irregular horse and formed 
perhaps the most gallant, certainly the most magnifi- 
cent body of irregular cavalry in the world. But they 
cherished few but personal objects. All the revenues 
that could be wrung from the country went to cover 
them splendidly in armour, to fill their stables with the 
noblest Arab stock, to adorn their palaces with the 


finest carpets of the East, to crowd their harims with 
beautiful slaves and to guard them with negro eunuchs. 
Under the rule of these picturesque but stupid warriors 
the resources of Egypt rapidly decayed. The canals 
on which cultivation depended choked themselves by 
neglect. The desert encroached, while the cities shrank. 
Alexandria from a great and flourishing port dwindled 
to a town of 5000 inhabitants. Wandering Bedouins 
perpetually raided the settled areas. Caravans from 
Suez or Kossir to Cairo could pass in safety only under 
a strong military escort. In fact what Sind was under 
the Mirs, Egypt was under the Mamelukes. 

The advent of the Ottoman Turks had led to the 
abandonment of the ancient trade routes by Baghdad 
and the Persian Gulf and by Alexandria and the Red 
Sea, which for ages had borne the bulk of the traffic 
between the East and West. But in the middle of the 
eighteenth century events in India called for swifter 
communications with Europe than could be provided 
by way of the Cape. The schemes of Dupleix, the 
achievements of Clive, the struggles ofWarren Hastings, 
the whole absorbing question whether India and the 
Indian trade was to be controlled from Paris or London, 
demanded prompt decisions and early reinforcements. 
Conditions in Egypt, Syria and ’Iraq became therefore 
matters of considerable interest to the rival western 
powers. From early times the English East India Com- 
pany, whenever it had had urgent news to send east- 
wards, had despatched couriers overland, usually by 
way of Aleppo and Baghdad, to take ship at the head 
of the Persian Gulf. But this route had presented many 
disadvantages. The growing disorder of the pashaliq of 
Baghdad, and the frequent depredations of the Arab 
tribes had rendered it very insecure. The Company’s 
packets contained nothing to excite Arab greed; but 
even if the tribesmen did not fancy the messenger to be 



well supplied with money, they might always be carried 
away by the sporting pleasure of slaying the infidel. 
Though many packets reached their destination, there 
was always a good chance that the bearer might he 
murdered or at least be obliged to destroy his papers. 1 

There was however an alternative route by way of 
Egypt and the Red Sea. This had the advantage of 
offering a shorter passage through nomadic areas — 
merely from Cairo to Suez; and, providing an agree- 
ment could be reached with the beys, it promised much 
greater regularity and security. When in 1768 the 
traveller James Bruce visited Egypt, he found Ali Bey, 
the virtual ruler of the country, in open revolt against 
the Turks and not at all unwilling to make friends with 
the infidel as an insurance against Turkish hostility. 
He was moreover enlightened enough to perceive that 
by encouraging trade he might enlarge his revenue. 
Bruce’s suggestions, backed by Italian merchants 
settled at Alexandria, were reinforced by the direct 
proposals of British sea captains, who found the Red 
Sea trade decaying and fancied that they might find a 
good market for their Bengal goods in the bazaars of 
Cairo. Ali Bey was sufficiently interested to address a 
letter to the English authorities in Bengal, suggesting 
that they should open a direct trade to Suez, in defiance 
of the Sultan’s orders that no Christian vessels should 
be admitted to any ports north ofjidda. 2 When Warren 
Hastings became governor of Fort William in 1772, he 
perceived at once the advantages that might accrue to 
Bengal from these proposals. Under his auspices more 
than one expedition was despatched, and a provisional 
treaty was actually signed by which Ali Bey’s successors 
guaranteed the safe passage of goods from Suez to 

1 See for instance the adventures of Captain James Barton 
(Madras Public Consultations, August 10, 1758). 

2 Cf. Charles-Roux, Autour d’tine Route, pp. 29 sqq. 


Cairo. 1 But these arrangements pleased neither the 
East India Company nor the Sultan, who had just re- 
covered something of his precarious power over Egypt. 
The Sublime Porte feared that the Hijaz revenues 
would suffer if the Indian trade were diverted from 
Jidda to Suez ; while the Company feared that the trade 
development of the Egyptian route would lead to a 
breach of its monopoly by the export of prohibited 
goods from India to Europe by way of the Mediterra- 
nean. So in 1777 the Company forbade the despatch 
of vessels with trading cargoes to any port north of 
Jidda, but at the same time obtained from the Porte a 
verbal promise of free passage through Egypt for its 
despatches and couriers. This of course had little effect. 
Neither Company nor Porte could fully enforce these 
orders. Packets were made a pretext for surreptitious 
cargoes; and in 1779 and again in 1780 bearers of 
English packets were seized and imprisoned. 2 

Meanwhile the French had shown themselves keenly 
alive to the possibilities of the Egyptian route. To them 
it promised an incalculable advantage — a way of mini- 
mising the English superiority at sea which had exerted 
such a distressing influence on the course of the Seven 
Years’ War. If the main stream of the Indian trade 
could be turned into the Mediterranean, not only 
would the opportunities and advantages of French mer- 
chants be greatly enhanced, but also the duties of the 
French navy would be greatly reduced. Such ideas too 
seemed the more timely in view of the evident decay of 
the Turkish empire. When it actually should collapse, 
that event would of course immediately benefit neigh- 
bours such as Austria and Russia; but, as a French 

1 Brit. Mus. Add. MSS, 29210, ff. 422 sqq. A copy of the treaty 
occurs in the India Office Factory Records, Red Sea, vol. 5 a. 

2 Charles-Roux, op. cit. pp. 124, 148. James Wooley, one of 
those concerned, had been commandant of the Nawab of Arcot’s 
artillery (Madras Public Sundry, No. 4). 



memoir of 1783 observed, their advantages might be 
easily and appropriately counterbalanced by a French 
occupation of Egypt. 1 An alternative plan, which had 
the merit of being immediately practicable, was that of 
entering into an alliance with the beys. This was done. 
Early in 1 785 a French agent signed agreements with 
the leading bey, with the chief customer, with one of 
the Bedouin chiefs, for the safe transit of French goods 
on favourable terms. 2 This arrangement however, like 
Warren Hastings’ provisional treaty, only served to em- 
phasise the uncertainty of the Egyptian situation. Not 
only did the Porte refuse to confirm the French treaty, 
but once more reasserted its fluctuating authority at 
Cairo. 3 For the moment, therefore, this danger to the 
English position in India vanished; but none the less 
it remained true that at any time the French might 
establish themselves in Egypt either by force or by 
negotiation. 4 

We therefore followed the French example. George 
Baldwin, who had played a prominent part in our 
earlier projects, was named consul-general, with in- 
structions to get from the beys as good a treaty as had 
been obtained by the French. The revival of Turkish 
influence rendered that harder than had been ex- 
pected. Year passed after year, and, as far as the 
Foreign Office could see, Baldwin was drawing £1400 
a year for doing nothing; Grenville resolved in 1793 to 
abolish his post, since if he was worth keeping in Egypt 
at all his pay should be found by the East India Com- 
pany. 6 This decision was promptly followed by news 
that at last Baldwin had got his treaty. But with the 

1 Charles-Roux, op. cit. p. 167. 

2 Idem, pp. 168 sqq. 

3 Hailes to Carmarthen, October 12, 1 786 (India Office, French 
in India, vol. 13). 

4 Eton, Survey of the Turkish Empire , p. 495. 

6 Dropmore MSS, n, 263, 273. 



exception ofDundas the government of the day had lost 
all interest in Egypt in view of the much more im- 
mediate danger arising from the revolution in France. 

The French themselves, however, soon dispelled our 
indifference. In the winter of 1797-8 a number of 
converging influences led them to plan a great ex- 
pedition eastwards. In the spring Grenville learnt that 
the state libraries had been ransacked for the works of 
travellers in Egypt, Persia and India; that the services 
of Arabic, Turkish and Persian scholars had been re- 
quisitioned ; and that the expedition was. said to be 
designed to occupy Egypt and cut through the isthmus 
of Suez. 1 It was not known how much of this could be 
taken seriously. But Dundas thought the project “a 
great and masterly stroke ” ; 2 and the Governor-General 
at Calcutta resolved to break or bridle Tipu Sultan be- 
fore Bonaparte’s “desperate and enterprising spirit” 
could find the means of joining him with a body of 
French troops. 3 In England it was resolved that, what- 
ever might be the purpose of the armament assembled 
at Toulon, every English ship that could be spared 
should be sent to disperse or destroy it. “ Never ”, wrote 
Mornington with great truth, “was a public measure 
taken with more wisdom or spirit.” 4 

Meanwhile Napoleon had sailed from Toulon with 
some 38,000 men on May 19. On June 12 Malta sur- 
rendered to him. At the end of the month he landed on 
the Egyptian coast near Alexandria, promptly occupied 
that city, and then marched southwards. On July 18 
he crushed the Mamelukes opposite Cairo at the battle' 
of the Pyramids, and entered the city on the 24th. Eight 
days later Nelson found the French squadron, which he 
had been hunting for weeks, and destroyed it in Abukir 
Bay. Then the throttling effects of superior sea power 

1 Dropmore MSS, iv, 192-3. 2 Wellesley Despatches, 1, 348. 

3 Idem, 1, 322. 4 Dropmore MSS, iv, 385. 


began to display themselves. Cut off from supplies, 
from reinforcements, and even from news on which he 
could rely, Napoleon might employ his matchless 
powers of organisation, creating a government, con- 
ciliating the religious leaders of Cairo, suppressing 
tumults, framing magniloquent proclamations, but, 
from a French point of view, he was merely ploughing 
unwatered sand. He sought to break out by way of 
Syria, but again the ships of his enemies carried to Acre 
supplies and men and a leader who frustrated the 
French efforts to take the place. He might announce to 
the citizens of Cairo that he had beaten down the walls 
of Acre and bombarded the city till there remained not 
one stone upon another, 1 but he was defeated. In 
sullen recognition of the fact he deserted his army in 
Egypt, and sailed for France on August 23, 1799. 
KJeber, who was left in command, with good reason 
disliked and distrusted his position. When a Turkish 
army advanced against him, he entered into negotia- 
tions with Sir Sidney Smith, the defender of Acre. By 
the convention of Al-Arish, signed on January 24, 
1800, it was agreed that the French should evacuate 
Egypt in vessels to be collected and supplied by the 
Turks. But, while the Turks were assembling ships in 
their usual leisurely manner, the English cabinet had 
time to decide, on a false estimate of the French posi- 
tion, that the convention should not be carried into 
effect. This costly blunder involved us in a special 
expedition to turn the French out of the country. At 
the close of the year Sir Ralph Abercromby was sent 
with 15,000 men for this purpose, while a force from 
India co-operated on the side of the Red Sea. Aber- 
cromby landed at Abukir Bay on March 8, 1801. By 
this time Kleber had been murdered, and the command 
of the French had devolved upon Menou, an incapable 
1 Jabarti, Merveilles Biographiques et Historiques, vi, 132. 


soldier who had espoused Islam and an Egyptian wife. 
A battle was fought outside Alexandria. Abercromby 
perished, but the issue was that one body of the French 
shut itself up in Alexandria while the remainder, 12,000 
men, manned the walls of Cairo. This unenterprising 
behaviour promised no obstinate resistance. First Cairo 
and then Alexandria surrendered, and the French 
occupation of Egypt came to an inglorious end. But it 
had been far indeed from fruitless. It had shaken 
Mameluke power; it had fully awakened English 
minds to the strategic importance of a country placed 
midway between East and West; it had illustrated 
Turkish incompetence ; and incidentally it had brought 
to Egypt an Albanian adventurer, Muhammad ’Ali. 

Muhammad ’Ali had been born in 1 769 at the tiny 
walled seaport of Kavala, in a little house in one of the 
old winding streets. Of the origins of his family nothing 
is positively known. Turkish and even Persian ancestry 
is claimed for him ; the first seems corroborated by the 
sturdiness of his physique and the fortitude of his 
character, the second seems to accord better with his 
subtle, flexible intelligence. In any case his family was 
undistinguished. His father, Ibrahim Aga, commanded 
a small body of local irregulars in the governor’s ser- 
vice ; but, dying while his son was still young, left him 
to the care of the governor. The boy’s education must 
have been severely practical. He was fed and clothed, 
he learnt his prayers, he was taught to ride, he was 
practised in the use of arms. As soon as he was old 
enough he must have accompanied parties sent out to 
capture bandits or get in the revenue, and so learnt the 
rudiments of war, the art of surprise, the manner of 
command. He is said himself to have conducted some 
of these parties with remarkable success, but here we 
are at the mercy of the anecdotist, who loves few things 
so well as to explain mature eminence by early traits of 


genius and to contrast present greatness with past 
humiliations. When the boy was eighteen, he was 
married to one of the governor’s relatives, who became 
the mother of five of his ninety-five children. He then 
embarked on the tobacco trade — the finest Turkish 
tobacco grew in the district lying inland of Kavala — 
but on what scale we do not know. Some have fancied 
him trading largely with a fortune derived from a 
wealthy wife; 1 others repeat stories of his having been 
helped out of a difficulty by a trifling loan of a couple 
of rubiehs . 2 But all we really know is that he himself 
looked back upon this early life with affectionate regret, 
towards the end of his days revisiting his birthplace and 
endowing a school there which still survives. 3 

When under British pressure the Porte assembled 
forces in the vain hope of expelling the French, the 
governor — Shorbashi was his Turkish title — of Kavala 
was called upon to furnish a contingent of 300 men. 
This party was sent under the command of the gover- 
nor’s son, ’Ali Aga, with Muhammad ’Ali as his 
lieutenant. But the passage to Abukir was stormy, and 
the Turkish force on landing had to bear many priva- 
tions before it was unceremoniously bustled back into 
the sea by the French. The story runs that in this 
hurried re-embarkation Muhammad ’Ali himself was 
nearly drowned, and was rescued by a British man-of- 
war’s boat. Be that as it may, the commander of the 
Kavala contingent was disgusted by the sea-sickness, 
hunger, and thirst that he encountered, and speedily 
returned home, leaving Muhammad ’Ali in command. 
The latter’s spirit and resource attracted the attention 

1 Sabry, V Empire igyptien, p. 24. 

2 Politis, L’Hellenisme et VEgypte rnoderne, p. 179. 

8 Letter to the Governor of Kavala, 9 Shawal, 1246 (Abdine 
Palace Records); and Stoddart, No. 7, August 29, 1846 (F.O. 
78-661 b). 



of the Turkish commanders, while his forethought and 
attention secured the confidence of his men; thus he 
became by 1801 one of the two chief officers in com- 
mand of the Albanian troops who formed the main 
strength of the Turkish force in Egypt. 

This body had co-operated with the English expedi- 
tion to the extent of plundering the open country and 
occupying the places not held by French troops. 
Hutchinson, Abercromby’s successor in the command, 

soon formed the poorest opinion of its value, and 
hravely doubted its ability to maintain itself in Egypt, 1 
and this view was confirmed by the request made by 
the Porte itself that a British force should remain in the 
country after the expulsion of the French, as a pre- 
caution against any renewal of their attempts. 2 The 
occupation of strategic points such as Alexandria and 
Suez, at all events till the end of the war, was suggested 
by the Russian ambassador 3 and advocated by the 
Governor-General of British India. 4 Pamphlets were 
'written in the same sense, 5 and Dundas, who had 
; ilways held strong views on the importance of Egypt, 
flavoured the idea. Hawkesbury’s ministry inclined to 
t|he same plan and was especially anxious to secure such 
a . settlement of the relations between the Sultan and 
t le beys as would prevent the recurrence of that mis- 
g overnment which had assisted the French occupation, 
v Vith this object it was proposed that the rights and 
d uties of the Mamelukes should be defined, that the 
c ollection of the revenues should be regulated, and that 
a fixed amount should be assigned to the maintenance 
off a military force under the command of British 

V Cf. Charles-Roux, U Angleterre et V Expedition franpaise en 
Efgypte, n, 262. Douin and Fawtier-Jones, La Politique Mimeluke, 

P| 5 ' 

y Charles-Roux, op. cit. n, 268. 3 Dropmore MSS, vn, 15-16. 

t Wellesley Despatches, n, 588. 

’I E.g. Baldwin, Political Recollections relative to Egypt. 



officers. 1 Unfortunately, Elgin, our ambassador at the 
Porte, was not the man to persuade Turks that a dis- 
agreeable or honest measure could be to their ad- 
vantage. Instead of making terms, they promptly gave 
a striking demonstration of their proverbial bad faith. 
The Capitan Pasha — as the Turkish admiral was called 
— lured a body of Mamelukes on board two barges, 
where they were suddenly fired upon, and those whc 
were not killed were made prisoners. This significant 
episode nearly led to blows between the Turk and 
English forces, and the Turks were only induced tc 
release their captives by threats. 2 The beys then retiree 
to Upper Egypt, out of Turkish reach. 

While these commotions and disputes had been going 
forward, the Treaty of Amiens had been signed, 
declaring that Egypt was to be restored to the Sultan' 
This gave the French the right to demand its prompt 
evacuation by the English forces. After a few faint- 
hearted attempts to settle the question of the beys, the 
English commander was satisfied by the issue of orders 
that the beys were to be pardoned and granted the 
province of Assouan. The troops were then embarked • 
one of the Mameluke leaders, Elfi Bey, accompaniet 
them on a visit to London; and Major Missett wa 
appointed English agent to watch over the behaviou 
of both Turk and Mameluke, and do everything in hi 
power to keep the French out of the country. Th 
English occupation thus ended in March, 1803. 

Missett’s appointment as English agent was no dout 
meant to counteract the intrigues of the consuls whor 
the French reappointed after the Treaty of Amier 
Similar agents were sent by the other Mediterrane; 

1 A good summary is given in Ghorbal, Beginnings of the Egypt 
Question , p. 166. The document is printed app. Douin and Fawti 
Jones, op. cit. p. 50. 

2 Douin and Fawtier-Jones, op. cit. pp. 160 sqq. 


powers, and later on by Sweden, Prussia and Russia. 
These European agents were sharply divided into two 
classes according as they devoted most of their time to 
trade or to political functions, and naturally this divi- 
sion tended to follow the relative importance of the 
powers whom they represented. In the early years of 
the nineteenth century political functions amounted to 
little. Early consuls-general like Salt or Drovetti spent 
as much time collecting antiquities as in representing 
their national interests. But from 1830 onwards their 
political duties developed a new importance, and they 
became in fact, though not of course in form, diplo- 
matic agents stationed with the pasha and seeking to 
guide his action in accordance with national policy. 
Some became close and intimate friends of Muhammad 
’Ali, and the personal influence of Colonel Campbell 
weighed much with him in determining his admini- 
strative, if not his external, policy. 

The English departure cleared the stage for as pretty 
a piece of skilful intrigue, plot and counter-plot, as ever 
vas brought to a triumphant conclusion. In appear- 
ance only two parties were seeking the possession of 
tgypt — the Turks and the Mamelukes. But in reality 
aatters were less simple. The Turks themselves were 
tjvided. On the one side were the troops who would 
gyey the orders of Khusrau Pasha, appointed by the 
\jltan to govern Egypt; and on the other were the 
djbanians indisposed to obey anyone except their own 
c«ef leaders, Tahir Pasha and Muhammad ’Ali. The 
aamelukes too had their own Bardissi and Elfi factions, 
ofch much more disposed to destroy the other than to 
ke common cause against a common danger. The 
^asible combinations were thus considerable, and the 
rational forecast that could have been made at the 
Jnent was that none of them would be likely to endure 


The first to disappear was Khusrau Pasha. When at 
a later time he had become Grand Vazir of the empire, 
western diplomatists regarded him as “a shrewd, bold, 
illiterate barbarian 1 But in 1 803 his character hardly, 
seems to have yet risen to these heights. He is described 
at this time as possessing no knowledge of war, politics 
or administration beyond the cutting off of heads. 2 He 
lay, of course, at a considerable disadvantage in Egypt. 
The Turks were there regarded with even more hatred 
than the French. Their ignorance and mispronuncia- 
tion of the sacred language, their pride, their claims to 
rule the country, helped to rob them of all local sup- 
port. “ God destroy them all ”, sighs the pious chronicler 
of the period. 3 Meanwhile at the head of the Albanian 
troops was Tahir Pasha, who had originally commanded 
a band of professed brigands in his native country with 
sufficient success and ferocity to merit admission to 
military rank in the Sultan’s army. In Egypt he had 
displayed much courage and resource, but had been 
disappointed of his promised reward. 4 His followers 
too were dissatisfied with their lack of pay. Accordingly 
in May, 1803, they mutinied at Cairo. This was c 
course a mere everyday incident in a Turkish arm; 
But when Tahir offered his mediation, Khusrau refuse 
it. The next day Tahir at the head of his Albania 
attacked and occupied the citadel. Khusrau fled 
Damietta, and Tahir assumed the government of t 
city. As he had not been joined in this movement 
the Osmanli troops, he called in the Mamelukes to j 
aid ; but before they could arrive, he was murdered. I 
death, however, made no difference to the immedi 
situation. His place was taken by Muhammad ’7 

1 Bulwer, Life of Palmerston , ix, 252. 

2 Douin, JJEgypte de 1802 d 1804, p. 18. 

* Jabarti, op. vn, 241. 

4 Leake, Observations upon Greece (F.O. 78-57). 



and the Mamelukes with the Albanians defeated 
Khusrau before Damietta, and finally conducted him a 
prisoner to the citadel of Cairo. Here then was the 
first combination: Albanians and Mamelukes against 
the Turks. 

When all this was known at Constantinople, another 
governor, ’Ali Pasha, was hurriedly despatched with 
1 500 men to replace the deposed Khusrau. 1 He reached 
and occupied Alexandria, but promptly embroiled 
himself with the consuls of the European powers 
established there. He publicly declared that no capitu- 
lations held good where he was governor. He exercised 
no control over his troops, who amused themselves by 
firing at the Swedish arms over the consulate. He even 
sought to interfere with a decision of the local court 
which had inexplicably gone in favour of the Frank. 2 
Early in 1804, expecting the assistance of the Albanians 
under Muhammad ’Ali, he moved southward against 
Cairo. 3 But the Albanians did not move. The pasha 
was captured and put to death by Bardissi Bey. 4 

He was succeeded by a third pasha, Kurshid. By 
this time the alliance of the Mamelukes and Albanians 
was wearing thin. The latter were exigent over their 
pay. The beys were reduced to forced loans and other 
violent expedients. It grieved them to the heart, to rob 
for other men’s benefit. Moreover they inclined to 
support Kurshid as pasha, by reason of his pacific and 
moderate character. 5 A new grouping was therefore 
indicated, and developed with all the certainty of 
predestination. Elfi Bey returned from England in 
February, 1804. Instantly the Bardissi faction, with 
the support of the Albanians and perhaps with the 
encouragement of Muhammad ’Ali, attacked Elfi’s 

jpouin, UHgypte de 1802 d 1804, pp. 55-6. 
idem, p. 1 13. 3 Idem, p. 143. 




followers, whose houses in Cairo were plundered. Mu- 
hammad ’Ali was delighted at this division of the beys, 1 
and promptly sought another ally, the new pasha who 
still lingered at Alexandria. He told the French agent 
at Cairo that as soon as the Albanians could get some 
pay, eight months in arrears, from the Mamelukes, 
there would be such an explosion as would restore the 
former to the Sultan’s favour. “What can we, thei 
natural enemies, expect”, said he, “from men so read 
to betray their own brethren?” 2 

The explosion followed in due course. On March : 
the Albanians at Cairo attacked the houses of the leadir 
beys. The citadel surrendered; and Muhammad ’A 
proclaimed the farmans appointing Kurshid Pasha Wz 
of Egypt. 3 The pasha naturally proceeded to join 1. 
ally, and there followed some months offighting arour 
Cairo, between the beys on the one side and the pasl 
and Muhammad ’Ali on the other. But whereas in tl. 
previous year Albanians and Mamelukes had had * 
unite to expel Khusrau, now Turkish power had falle 
so far that Kurshid was dependent on Muhammad ’Al 
who was visibly waxing in authority. The pash; 
Lesseps reports truly enough, was nothing but a 
instrument which the Albanians could use at will. 4 1 
the autumn of the year this fact was exhibited in a mo 
striking manner. The Albanians were weary of Egyj 
and many desired to return home with the plunr 
they had obtained. But Kurshid felt that he could < 
maintain his position without the cool, resolute 
dexterous aid of Muhammad ’Ali, and induce* 
latter, we must suppose without much difficul 
remain. 5 The misery of Cairo at this period was ext 
Muhammad ’Ali’s troops had to be satisfied 



Douin, L'Egypte de 1802 a 1804, p. 173. 

Idem, p. 180. a f demj p . 




Kurshid had to renew for their benefit the exactions 
formerly enforced by the Mamelukes for the same 
altruistic purpose. The leading Copts, for instance, were., 
led up into the citadel and requested to provide 2000 
purses. 1 The Mamelukes ranging round Cairo cut off 
supplies of grain so that famine raged in the city. Good 
Muslims began to regret the infidel government of the 
French. 2 

The foreign agents in Egypt contemplated these 
events with the normal human inability to forecast the 
future. A couple of generations later, with a view to 
stimulating Khedivial generosity, a story was put about 
that Lesseps, the French agent, had early recognised 
the genius of Muhammad ’Ali and contributed to his 
rise by his countenance and advice. In striking con- 
trast with this romantic tale is his actual language to 
Talleyrand: Je ne crois pas qu’il ait le genie de concevoir un 
plan vaste et les moyens pour Pexecuter . 3 Accordingly he 
favoured and encouraged, not this leader of limited 
genius, but the Mameluke beys, whose return to power, 
he thought, would be favourable to the extension of 
French influence. 4 The English agent, Major Missett, 
suffered, excusably enough, from the same delusion. 

The events of 1805 were however to illuminate the 
situation. Starved by the Mamelukes and plundered 
by tfie pasha, the Cairenes grew restive in their misery 
and| turned to the Albanian leader for rescue. This, 
ho mver, was the result less of instinctive recognition 
than of discreet instigation. Muhammad ’Ali, so the 
Aral| chronicler of the time assures us, made friends 
secretly with one of the chief shaikhs, visiting him, 
flattening him, declaring that if he were the ruler of 

1 Af ‘purse” was a sum of 500 piastres, the normal unit of large 
transactions, like the bag of 500 pagodas in south India. 

2 Dduin, U£gypte de 1 802 d 1 804, p. 248. 3 Idem , p. 1 73. 

4 Id&pi, pp. 226-7. 


Egypt he would govern with justice and follow the 
advice of the religious leaders. 1 He thus began to 
secure a following in the city itself, which Kurshid tried 
in vain to control by holding as hostages two shaikhs, 
but not Muhammad ’Ali’s friend. 2 At the same time he 
sought and obtained reinforcements from Syria, to 
render him less dependent upon Albanian support. 
Their arrival brought matters to a head. Their com- 
mander proved to be a brother of one of Tahir Pasha’s 
murderers. 3 They showed themselves to be less dis- 
ciplined than either Albanians or Mamelukes. Terrible 
stories spread abroad of how they drove the citizens 
from their dwellings, raped and murdered women, and 
stole children. These stories, to judge from the like 
elsewhere, lost nothing in the telling. But their probable 
exaggeration in no way diminished their moral effect. 
The whole city of Cairo fell into panic. The gates of the 
great mosque of A 1 Azhar were closed; the bazaars 
and caravansarais were shut up. No man ventured out 
without feeling that he carried his life in his hands. 4 

At the moment of the arrival of this alarming garri- 
son, Muhammad ’Ali had been absent from Cairo 
campaigning against the beys. But he speedily returned 
and a week later entered the city at the head of 4000 
men. 6 His pretext was the demands of his men for their 
pay, demands with which the new-comers, like good 
Turkish soldiers, felt a lively sympathy. On May 9 
Kurshid, quite unable to interpret portents, too^ ad- 
vantage of Muhammad ’Ali’s return to read pu,blicly 
the farman conferring on him the pashaliq of Jidda. 
This flattering hint that his presence was no longer 
desired at Cairo was enough for the Albanian leader. 
As Kurshid was preparing to return to his residence in 

1 Jabarti, op, cit. vm, 70. j 

2 Douin, Mohamed Ali Pacha du Caire, p. 14. 3 Idem, jp. 1 1 . 

4 Idem, pp. 19-20. 3 Idem, Jo. 14. 


the citadel, he was suddenly surrounded by Albanians 
clamouring for their pay, accusing him of having made 
away with the public revenues, and threatening him 
with instant death. One of the Albanian officers came 
forward to protect him from actual violence; but the 
populace, under the guidance of the shaikhs, were 
already proclaiming Muhammad ’Ali governor of 
Cairo . 1 Kurshid succeeded in escaping back into the 
citadel, from which he tried to bombard the city into 
submission. But Turkish gunners were unable to achieve 
that task, and rather excited than alarmed the in- 
habitants. The shaikhs, encouraged by Albanian sup- 
port, put forward a great number of demands. Then 
as now it was regarded as sound practice in political as 
in commercial bargaining to begin with asking what 
you never expect to get. On this occasion they required 
that the troops should in future be stationed on the 
other side of the river at Ghiza, that no soldier under 
arms should be allowed to enter the city, and that no 
contributions should be levied on the citizens . 2 These 
being rejected, they again and more formally proclaimed 
Muhammad ’Ali as governor, and attempted to be- 
leaguer the citadel. Popular enthusiasm was immense. 
French observers were reminded of the heady zeal 
with which Parisians had flung themselves into the 
Revolution , 3 and indeed there was this similarity that 
in both cases the People were busy changing one master 
for another. But there was a fundamental difference 
too. The people behind the Parisian mob did indeed 
aim at new institutions; the man behind the Cairene 
tumults was at the moment seeking nothing but per- 
sonal power. The French stormed the Bastille. But the 
Cairenes, though they murdered Kurshid ’s men readily 
enough when they recognised them in the streets, and 

1 Demin, Mohamed Ali Pacha du Caire 9 p. 22. 

2 Idem> p. 27. 3 Idem , p. 35. 



though everyone, even the children, purchased arms, 1 
did not capture the citadel of Cairo. Muhammad ’Ali 
made a show of helping them. He had cannon dragged 
up the Mokattum Hills commanding the fortress, an.d 
sharpshooters were posted in the minar of the Sultan 
Hasan mosque. But the leader himself was in no haste 
to bring the matter to a violent issue. It would perhaps 
have cost him many men; his followers were not too 
reliable; above all he preferred to become Pasha of 
Cairo with the assent of Constantinople rather than as 
an open rebel against the Sultan. The French agent 
Drovetti, a far more acute observer than his predecessor 
the tavern-keeping Lesseps, 2 went to the bottom of the 
matter in a report which about this time he despatched 
to Paris. “The measures of the enterprising Albanian 
leader ”, he writes, ‘ Tnake me think he hopes to become 
Pasha of Cairo without fighting and without incurring 
the displeasure of the Sultan. Every act reveals a 
Machiavellian mind, and I really begin to think he 
has a stronger head than most Turks have. He seems 
to aim at obtaining power through the favour of the 
shaikhs and the people, so as to reduce the Porte to the 
necessity of giving him freely the position which he will 
have seized,” 3 

The issue was much as he anticipated. In June an 
envoy from the Sultan arrived at Alexandria with 
orders to confer the pashaliq on Kurshid or Muhammad 
’Ali, whichever seemed the stronger. After some delay 
the envoy recognised Muhammad ’Ali, and on August 
7 Kurshid quitted the citadel of Cairo and marched 
down to Bulaq to take boat for Alexandria. , 

The political dexterity displayed by Muharpmad 
’Ali in the course of these events was extraordinary. He 

1 Douin, Mohamed Ali Pacha du Caire , p. 35. 

2 Missett, September 3, 1804 (W.O. 1-347). 

3 Douin, op. cit. p. 35. 



had aided the Mamelukes to overthrow Khusrau Pasha ; 
he had then aided one Mameluke faction against 
another; he had then helped Kurshid against the 
Mamelukes; and lastly he had headed the Cairenes 
against Kurshid — thus weakening in turn both Turks 
and Mamelukes alike, but carefully keeping himself in 
the background, never committing himself over-deeply 
to any of the conflicting factions, and at last securing 
for himself the sanction of the Sultan’s authority. Some 
have seen in this a desire to legitimise his power. But 
in fact he was much too sturdy a political realist to 
place excessive value on mere abstract right. As a 
matter of fact, the Sultan’s recognition did little to 
strengthen his position within Egypt itself. He could 
expect neither an accession of troops nor even consistent 
moral support from Constantinople. The corrupt and 
improvident Divan would turn against him as soon as 
some stronger or more promising candidate appeared, 
and the Mamelukes still occupied the whole of Upper 
and much of Lower Egypt. But the imperial recogni- 
tion did promise a temporary respite from Turkish 
interference. For a few months at all events he would 
only have to face the Mamelukes, and not to balance 
precariously between Mameluke and Turk, unless by 
chance one of the European powers intervened. 

In any case the maintenance of his position remained 
most uncertain. His army could be held together only 
by pay or plunder; and it was therefore clear that he 
would be driven into the same course of exactions as 
had ruined his predecessors. Then too what were foreign 
powers going to do? Drovetti might recognise the new 
pasha’s strong sense and Machiavellian talents, but he 
did not at this time desire the continuance of his ad- 
ministration. Neither did the English representative, 
Missett. Both in fact distrusted either his good will 1 
1 Missett, January i, 1806 (F.O. 24-2). 



or his ability to maintain himself; 1 and both therefore 
set to work to encourage rival Mameluke factions. The 
joint hostility of both French and English thus helped 
Muhammad ’Ali forward, by making sure that the 
Mamelukes did not unite against him. Yet Drovetti at 
least lay under no illusions regarding the military forces 
of the beys. “All the chiefs combined”, he writes, 
“have not more than 800 Mamelukes; the remainder 
are a mob of Greeks, Osmanlis and Arabs, attracted 
to their tents by hopes of pillage. The Mamelukes are 
no longer the brave soldiers ready to follow their 
masters to the death ; they have no longer organisation 
or discipline. A bey’s court, once a school of military 
discipline and morals, has become the source of debauch 
and disobedience. Their wandering life of brigandage 
has degraded them.” 2 He concluded that Egypt would 
never know order or good government until the French 
occupation was restored. 3 

The English attitude was in most respects the con- 
verse. All our experience in the course of the expedi- 
tion of 1801 had led us firmly and rightly to the 
conclusion that the Turks probably could never re- 
establish and in any event could not maintain their 
power in Egypt. General Hutchinson described them 
as deplorably feeble, distrustful of their friends, relying 
on their enemies, lacking the talent to form plans and 
the energy to carry them into effect. 4 At the same time 
every one believed that the French still meditated the 
reconquest of Egypt. As soon as the war with France 
was renewed, Nelson, commanding in the Mediter- 
ranean, was warned to be on the watch for any expedi- 
tion eastward. For the same reason it had been re- 

1 Douin, Mohamed Ali Pacha du Caire, p. 99. 

2 Idem, pp. 82-3. 

3 Idem, p. 139. 

4 Douin, La Campagne de 1807, p. vi. 



solved to retain possession of Malta. But if the French 
meant to renew their attacks on Egypt, and the Turks 
could not defend it against them, the Mamelukes alone 
remained as the possible basis of an effective local 
government. Hence arose various projects for inducing 
the Porte to confide to the beys the administration of 
the country. When it became obvious that the Porte 
was resolved not to enter into such arrangements, 
projects for an English occupation, at least of Alexan- 
dria, emerged ; and since the Porte was equally dis- 
inclined to consent to this measure either, the English 
ministry began to entertain the idea that they might be 
compelled by French action to occupy Alexandria 
whether the Sultan approved or not. 1 These projects 
were confirmed when the Sultan, as a result of Napo- 
leon’s great continental victories of 1805-6 and the 
break-up of the coalition, recognised Napoleon as 
emperor and received his ambassador with unusual 
honours. This was interpreted as opening Egypt to the 
French, and it was thereon resolved to occupy Alexan- 
dria. After some delay a body of English troops, de- 
tached from the army in Sicily, occupied the town on 
the night of March 20-21, 1807, 2 and Missett at once 
summoned the Elfi group of Mamelukes to our assist- 
ance, while the French consul fled to Cairo and did his 
utmost to hasten measures of defence against the 

The invasion had been facilitated by the fact that 
Alexandria was not at this time under even the nominal 
authority of Muhammad ’Ali. In the course of 1805, 
at Missett’s suggestion, the inhabitants of Alexandria 
had sought and obtained a farman placing the city 
under the command of a naval officer wholly indepen- 
dent of the Pasha of Cairo, and though the latter had 
sought to bribe the commandant into admitting an 

1 Douin, La Campagne de 1807, p. xxi. 2 Idem, pp. 23, 31. 


Albanian garrison, Missett had succeeded in inducing 
him to reject the offer. 1 

On March 29 a British column 1400 strong set out to 
capture Rosetta. Its object was two-fold — to facilitate 
the entrance of provisions into Alexandria and by a 
resounding success to encourage the Mamelukes to fly 
to our support. But the enterprise was ill-planned and 
ill-conducted. In the first place Fraser, the General 
commanding, should have marched in person with the 
main body of his troops. 2 In the second place the de- 
tachment itself was ill-led. Wauchope, the officer in 
charge, employed the whole of his troops in the attack, 
without keeping any reserve ; and when he was killed in 
the assault, the next senior officer was too badly 
wounded to be present in person with the troops but 
not so badly as to be compelled to hand over the 
command. The result was that when the Albanian 
garrison had been driven out, no one posted guards to 
prevent their return or took measures to reassemble the 
troops, scattered and disordered by the assault. It was 
the story of Patna in 1763 over again. The enemy, 
finding they were not pursued, rallied; finding no 
precautions taken, re-entered the town; and finding 
our men off their guard, attacked. In the scene of 
confusion that followed, alarming messages reached the 
wounded general, who ordered the retreat to be 

This misadventure developed into a major disaster. 
As has always happened in the East, the smallest 
portent of changing fortune transformed the whole 
attitude of the people. When the news of the occupa- 
tion of Alexandria reached Cairo, a panic ensued. The 
Albanian troops there meditated nothing but flight into 
Syria. They paid twice the usual price for horses, mules 

1 Missett, January x, 1806 (F.O. 24-2). 

2 Diary of Sir John Moore, u, 167. 



and asses, to carry away their effects, and bought 
Venetian sequins for 14 piastres which in the ordinary 
course passed for 10. The fellahin themselves were 
ready to rebel, refusing provisions to parties of Muham- 
mad ’Ali’s troops not strong enough to help themselves, 
and in some places even massacring them. 1 But the 
failure at Rosetta entirely changed the situation. The 
Albanians recovered their lost courage. The Mame- 
lukes, instead of hastening down to join the English, 
first hesitated, and then came to terms with Muham- 
mad ’Ali. The populace sank back into submission, and 
the English could no longer get intelligence of the 
enemy’s movements. 2 

These developments permitted Muhammad ’Ali to 
assemble troops and send them northward against the 
English, who in April had made a further attempt upon 
Rosetta. This time they proceeded more cautiously, 
bombarding the place, but deferring any actual attack 
until the arrival of the Mamelukes promised by 
Missett. However Muhammad ’Ali’s troops appeared 
instead of our supposed allies. The besiegers were 
unexpectedly caught between two fires. After pro- 
longed and confused fighting, in which we lost 400 
killed and 400 more taken, our troops withdrew once 
more to Alexandria. 3 

Muhammad ’Ali, at Cairo, now followed the same 
temperate policy that he had employed against 
Kurshid. The ordinary Turk would have been puffed 
up by his success, would have killed or circumcised his 
prisoners, and would have hurried on to try and push 
the survivors into the sea, regardless of consequences. 
The pasha so far condescended to custom as to permit 

1 Douin, La Campagne de 1807, pp. 179-80. 

2 Cf. Missett to Isaac Morier, May 2, 1807 (I.O., Egypt and the 

Red Sea, 6). 4 

3 Cf. Douin, op. cit. pp. 73-82. 


the heads of the fallen to be paraded through the 
streets; but he did not allow himself to forget that 
sooner or later peace would have to be made, that 
British fleets would suffer few enemy vessels to enter or 
leave the port of Alexandria, and above all that Britain 
dominated not only the Mediterranean in his front but 
also Indian waters in his rear. He therefore treated his 
prisoners well, and in May sent one of them down to 
Alexandria with a confidential interpreter to discuss 
the terms on which the English would withdraw. He 
offered in return for the retirement of the expedition to 
release his prisoners of war, to protect British trade, and 
to oppose any European force that should seek either to 
occupy Egypt or to pass through it towards India. 1 
These proposals were for the moment rejected. But the 
Portland ministry which took office in the spring of 
1807 held much sounder strategic views than Lord 
Grenville’s ministry which it displaced. It was resolved 
to abandon Alexandria, which could be reoccupied at 
any time if necessary. On September 14 a convention 
was signed by which Alexandria was to be surrendered 
to the pasha, the English prisoners freed, and a general 
amnesty accorded to all who had assisted the English 
forces. 2 

So another crisis passed. Had it been well com- 
manded, the expedition might have led to the destruc- 
tion of Muhammad ’Ali’s rising power and the 
restoration of Egypt to the Mamelukes or the Turks. 
But it was ill-conceived and mismanaged, like our 
abortive first expedition up the Tigris in the last war. 
By an extraordinary oversight it was not accompanied 
by any officers who had served under Abercromby 
and Hutchinson in the former occupation; nor was it 
possible to detach sufficient men from the force in 

1 Douin, La Campagne de 1807, p. 113. 

2 Idem, p. 164. 


Sicily to achieve the object of their despatch. In all 
respects it seemed a complete and costly failure. But 
perhaps the failure was not quite so complete as it 
appeared, for Muhammad ’Ali’s sagacious mind drew 
sound conclusions from the episode. He judged that 
the French army was a much more remote instrument 
of power than the British navy, and began to reflect 
that Great Britain might be a valuable ally in the 
accomplishment of schemes which he was already 
beginning to contemplate. 

Thus little by little the situation in Egypt had begun 
to simplify itself. The Albanian leader was now the 
representative of Constantinople; French intervention 
was impossible; English intervention had failed. There 
remained only the Mamelukes to be reduced before 
Muhammad ’Ali could be regarded as the sole master 
of Egypt. But there was nevertheless a growing danger 
that the victor, like many a successful claimant to an 
estate through the processes of law, would find himself 
the heir of little but immoderate debt. The country 
itself was going fast to ruin. Upper Egypt “groaned 
under the iron sceptre of the Mamelukes”; Lower 
Egypt was “utterly unable to support the united 
numbers of the troops and of its population. To satisfy 
the demands of the government and of the more 
oppressive extortions of its agents, the husbandman has 
been reduced to the necessity of parting even with his 
implements of agriculture; most of the villages of the 
districts of the sea have been deserted ; the banks of the 
Nile, formerly so luxuriant, are now condemned to an 
unnatural barrenness ” . 1 

The maintenance of the army was, as it always had 
been and as it was long to continue, “the necessary 
cause of great disorders”. 2 In 1809, the pasha had 

1 Missett, January 1, 1806 (F.O. 24-2). 

2 Douin, La Campagne de 1807, p. 138. 


some 10,000 effectives in his service, but, according to 
Turkish custom, they drew pay and allowances for 
30,00c. 1 Their pay and the other inevitable expenses of 
government greatly exceeded even in 1806 the land 
revenue of Lower Egypt and the customs dues on a 
trade that had almost vanished. The deficit could be 
made good only by a renewal of those levies which had 
been formerly made by the Mamelukes and Kurshid 
Pasha, but which were now more difficult and more 
hated than ever, because almost every one had already 
been stripped of his movable property. Even Euro- 
peans were obliged to contribute, and even consuls 
were compelled to assent. 2 The Arab annalist describes 
the troops as being the only people left with money to 
lend, 3 and indeed the Albanians at this moment in 
Egypt occupied much the same position as Arab 
mercenaries came to do at Baroda or Hyderabad. The 
embarrassment of the situation lay in either having to 
pillage and provoke the population or to displease and 
annoy the troops. 

Muhammad ’Ali had indeed done his utmost to 
escape from this dilemma. He had, for instance, attached 
the shaikhs and chief men of Cairo to his cause by 
bestowing on them villages formerly belonging to the 
Mameluke beys, so as to separate the Cairenes from 
their former rulers in case their oppressions should be 
forgotten. 4 But, even so, difficulties broke out from 
time to time, and various leaders, or would-be leaders, 
had to be arrested. 6 More dangerous still was the 
attitude of his troops. As he was returning one day in 
October, 1807, from the citadel, a party of soldiers 
opened fire on him from a house as he rode by, 

1 Driault, Mohamed Ali et Napoleon, p. 34. 

2 Douin, La Campagne de 1807 , p. 19 1. 

3 Jabarti, op. cit. vm, 182. 

4 Douin, op. cit. p. in. 5 Idem, p. 137. 


wounding his horse and one of his companions. 1 A few 
days later some 500 Albanians and Osmanlis assembled 
before his house in the city and actually fired into his 
windows; and the situation was alarming enough for 
him to retire from the city into the citadel. 2 

Clearly a financial surplus instead of a deficit, to be 
attained by any means that could produce such a 
miracle, was the prime condition of any improvement 
in his position. One financial resource which occurred 
to him was trade. This was no very original idea. 
Those writers who have described the East India 
Company as despicable in eastern eyes because it 
traded have strangely misapprehended the position. 
The ordinary merchant was despised, not because he 
was a trader, but because he was defenceless. All over 
the East from Constantinople to Achin and Bangkok 
you would have found great nobles, rulers of provinces, 
sons and mothers of ruling kings, even emperors them- 
selves, personally and directly interested in commerce. 
To Muhammad ’Ali, who had traded in tobacco 
before ever he had betaken himself to that life of high 
crime which in the East stood for politics, the step was 
natural, obvious, unquestionable. 3 

In this fortune favoured him. The only people with 
whom he could hope to do good business were the 
English. During the later years of the Napoleonic war 
the French flag practically vanished from the Levan- 
tine seas. A French vessel entering the port of Alex- 
andria in 1808 is said to have been the first to do so for 
five years, 4 and another in 181 1 as the first for a year 
and a half. 6 French ships could not be insured at 
Marseilles under 50 per cent, premium, and no news- 
papers reached Alexandria but abominable Malta 

1 Douin, La Campagne de 1807, p. 207. 2 Idem , pp. 209-10. 

3 Driault, V Empire de Mohamed Ali (1814-23), p. 205. 

4 Idem , p. 9. 6 Idem , p. 137. 


Gazettes, reeking, as Drovetti said, with libels on the 
French government. 1 But at the same time the English 
were eager customers for grain. The victualling of their 
squadrons keeping watch and ward over the Medi- 
terranean from Malta and Gibraltar, the victualling of 
their ever-growing forces operating in the Peninsula, 
called for constant large supplies of wheat, and those 
years were seasons of scarcity except in Egypt where 
the Nile rose high and harvests were plenteous. 2 The 
pasha seized this heaven-sent opportunity. In true 
oriental style — the matter could be matched scores of 
times in the dominions of both the Moghul Emperor 
and the Ottoman Sultan — the export of grain became a 
monopoly, which at times was said to yield him a gross 
profit of 500 per cent. 

Drovetti, the French consul at Cairo, did his utmost 
to check this growing connection. But the only satis- 
faction he could get consisted of assurances that the 
pasha was seeking his own interests only, and hints 
that the English might be providing money and 
munitions to be used against themselves. 3 However, 
the traffic was not limited to the sale and purchase of 
grain. That was paid for, partly in bullion, partly in 
munitions, but also, it would seem, partly in English 
commodities. English watches, we read, sold far better 
than the Geneva watches (often fraudulently marked 
with English names) which the French had been 
accustomed to sell in Egypt; and a genuine Prior would 
fetch twice as much as a French watch of equal 
quality. 4 Printed cottons too were imported and used 
so largely as to displace the locally made article. 6 

What was worse, from the French standpoint, was 

1 Driault, L’ Empire de Mohamed All (1814-23), p. 26. 

2 Ghorbal, The Beginnings of the Egyptian Question, p. 281. 

3 Driault, op. cit. p. 1 1 7. 

4 Idem, p. 99. 3 idem^ p . ^9. 


that political friendship sprang out of this commercial 
intercourse. At first Drovetti’s advice and intrigues 
during the English expedition of 1807 had borne some 
fruit. He had demanded, for instance, and seems to 
have secured the sequestration of English goods im- 
ported under the flag of Jerusalem, 1 and waged 
successful war against an audacious monk who had 
ventured to publish at Alexandria the reported excom- 
munication of Napoleon. 2 But when in 1811 a French 
privateer put in to sell prize cargoes, and when in 1812 
another tried to sell an English vessel she had taken, 
the English agents offered a successful resistance. At 
their request the matter was suspended until they had 
obtained farmans from the Sultan, once more at peace 
with England, forbidding the sale of prizes or prize 
goods in Turkish ports by either of the combatant 
powers. But this, Drovetti observed bitterly, did not 
prevent the importation of prize goods condemned by 
the English court at Malta, while the French had small 
chance of reprisals. “What is the use”, he asks angrily, 
“of making prizes if you can’t sell them anywhere?” 3 
Trade, cast into the scales, was evidently tilting the 
political balance heavily in the English favour, the 
more so as it was so rapidly replenishing the pasha’s 

At the same time and with the same purpose Muham- 
mad ’Ali began to overhaul the revenue machinery. 
In the Turkish, as in the Moghul empire, political 
decay had followed the same course. On all sides 
public revenue had been diverted to private advantage. 
The modes of assessment and collection had been so 
elaborated as to perplex and confound enquiry. The 
Copts, who had long monopolised the business of 
public accountancy, had developed accounts as com- 

1 Driaul t, U Empire de Mohamed Ali (1814-23), pp. 18, 20, 27. 

2 Idem, p. 63. s Idem, p. 191. 


plicated as those of the Brahmans at the Poona Daftar. 
A variety of coin gave opportunity for fleecing both the 
peasant and the government. The feddan, like the 
Indian bigha, did not mean the same area of land in 
different districts, or even in different parts of the same 
district. The practice of keeping officials in arrears of 
pay gave them an excuse (for what perhaps they would 
have done in any case) to impose and levy extra dues, 
which, as in India, were when discovered added to the 
public demand and at once replaced by additional 
impositions. Muhammad ’Ali resolved to cut his way 
through this forest of abuses. In 1808 he began an 
enquiry into landed tenures. Every oriental reformer 
has done so and been grievously abused for his pains; 
and what Akbar undertook in India, the pasha under- 
took in Egypt. It was indeed necessary. The pressure 
which he was putting on the revenue officials was not 
making them disgorge their perquisites but merely 
press more heavily on the fellah, who in consequence, 
weary of being plundered by the multazim (or zamin- 
dars, as they would have been called in India), by the 
officials, by the Bedouin, by the Mamelukes and by the 
Albanians, was abandoning his land and leaving his 
fields un tilled. Muhammad ’Ali therefore ordered an 
examination of all grants by which the multazim claimed 
to hold lands, and he then annulled all which were 
reported to be irregular, and a little later expropriated 
those who were in arrears of revenue, allowing pensions 
to the persons thus dispossessed. Some six years later, 
he carried the process farther, abolishing all the 
immunities till then enjoyed by all religious endow- 
ments — wakuf ; ordering the remeasurement of lands, 
when many were found paying revenue — rnirv — on only 
half their cultivated areas; simplifying the mode of 
assessment; and finally in 1814 expropriating the 
surviving multazim. These measures were severe, and 


most unpopular with the persons to whom Muhammad 
’Ali owed his simulacrum of popular support. But 
some such measures were urgently needed. Drove tti 
declared that in 1808 two- thirds of the land cultivated 
in 1798 was lying waste. The land which the pasha 
thus took into his own hands (putting into practice the 
theory of the East India Company’s servants) was not 
allowed to lie fallow. Fellahin were set to till it, under 
severe penalties if they neglected their task. 1 This 
interference with “the rights of property” which the 
English Whigs could never forgive, passed, not with 
that universal reprobation commonly asserted, but 
with only a few small gatherings in A 1 Azhar mosque, 
producing nothing but promises of alleviation which no 
one expected to be kept. 2 

These financial operations gradually re-established 
the treasury at Cairo, and the pasha’s own army 
became a less formidable menace in proportion as it 
could be regularly paid. 

At the same time the Mameluke question was 
gradually brought to a decisive settlement. In 1807, as 
we have seen, the beys had been induced by the 
intrigues of Muhammad ’Ali and Drovetti, by their 
own dissensions, and by the English failure at Rosetta, 
to neglect the last chance that fate was to offer them of 
re-establishing their power at Cairo and in Lower 
Egypt. But they still formed a very dangerous body, in 
occupation of Upper Egypt, successively threatening 
Cairo, or driven southward, according to temporary 
shiftings of military superiority. From time to time also 
there would be sudden negotiations, occasionally 
issuing in agreements which neither side meant to 
keep, and which only lasted about as long as they had 

1 Jomard, Coup d’CEil, p. 1 1 ; Driault, V Empire de Mohamed Ali 
(1814-23), pp. 231, 241 ; Jabarti, op. cit. vm, pp. 343, 345. 

2 Paton, Egyptian Revolution , n, 27; Driault, op. cit. p. 242. 




taken to negotiate. 1 The survivors of the Elfi faction 
still cherished hopes of a new and more powerful 
English expedition to overthrow their enemy and then 
go away again in their ships; and the more sanguine 
fancied that they might extract a large sum of money 
from the English to enable them to buy Muhammad 
’Ali’s troops and then overthrow him themselves. 2 The 
pasha, on the other hand, was bent on their complete 
subjugation, and aimed at inducing them to return to 
dwell at Cairo under his authority. Alternate negotia- 
tions and campaigns thus occupied many months after 
the departure of the English; and at last, towards the 
end of 1809, the beys did agree to come down and 
settle at Ghiza. 3 They did not arrive for nearly six 
months, and when they did come, they came far more 
prepared for war than peace. For some time their 
force and Muhammad ’Ali’s lay facing each other ; and, 
though a group of the Mameluke leaders decided to go 
over to the pasha, the majority resolved on a renewal 
of the conflict. A series of actions followed, in which 
Muhammad ’Ali’s artillery assisted him to success ; and 
at last about the beginning of 1811 a majority of the 
survivors decided to make their submission. 4 Their 
power in fact was broken. 

Time-honoured policy now demanded that they 
should be destroyed. Every good Turk held (with 
honest Pym) that “stone-dead hath no fellow”; and 
Muhammad ’Ali resolved to make an end of these 
fallen tyrants of the country. To achieve this it was 
desirable to assemble as many as possible in some 
secure place from which they could not escape. On 
March 1 the pasha’s son was to be invested with a dress 
of honour as Pasha of Jidda and commander of the 

1 Driault, UEmpire de Mohamed Ali (1814-23), p. 33. 

a Idem, p. 43. * Idem, p. 54. 


troops to be despatched against the Wahabi heretics in 
the Hijaz. All the Mameluke chiefs were invited to the 
ceremony, and encouraged to bring with them as many 
followers as they chose. Entirely deceived, they thronged 
to the citadel, to take part in the procession which was 
to march to camp by the Gate of Victories — the Bab-al- 
Futuh. From the platform of rock on which the chief 
buildings of the citadel were erected there ran down to 
the Bab-al-Azab (by which you pass to the Maidan 
Rumeila) a steep winding passage, cut in the rock, and 
commanded at every point for the destruction of any 
enemy who should force the gate. Down this path the 
troops moved — first some bodies of Ottoman soldiers, 
then the Albanians, then the Mamelukes, and then 
another body of infantry and horse. But when the 
foremost party of the troops had reached and passed the 
gate, the Albanian commander ordered the gate to be 
shut and barred, and his men turned and opened fire on 
the descending Mamelukes. The passage was speedily 
blocked by dead men and horses. Those who survived 
were either shot as they sought helplessly to escape or 
were seized, carried before the pasha, and beheaded by 
his orders. One man alone is said to have survived. 
Nor did even this massacre terminate the business. 
Troops were instantly despatched to slay the remaining 
Mamelukes wherever they might be found. The beys’ 
palaces were sacked. One European visitor had gone 
to a house near the citadel to watch the procession. 
Hurriedly returning home, he saw several of the 
unfortunate prisoners being led to the slaughter; one 
was cut down close beside him; he met the women of 
one of the beys being driven along by a party of 
Albanians like a flock of sheep ; and everywhere he saw 
soldiers loaded with plunder and drunk with fury . 1 

About a year later the pasha succeeded in bringing 
1 Lane-Poole, Life of Stratford Canning, 1, 107-9. 



off a second coup of the like nature. A number of 
Mamelukes still remained in Upper Egypt. After 
being chased and harried for some months by a force 
under Muhammad ’Ali’s son, Ibrahim, some 800 with 
200 of their slaves surrendered, and were put to the 
sword. 1 These Cromwellian measures left the pasha for 
the first time undisputed master of the country. 

Drovetti’s comments on these extraordinary events 
are not a little illuminating. On March 4, when pools 
of blood were still lying in the citadel and the city still 
bore the marks of the sacking of the beys’ palaces, it 
seemed to him a “terrible execution” which had 
deprived the English of their few remaining friends. 2 
But when Missett dared to add his congratulations to 
those of the French, and when it became apparent that, 
so far from having been impaired, the position of the 
English was steadily improving, the French consul 
belatedly awoke to the moral aspect of the tragedy, and 
the later measures against the Mamelukes are charac- 
terised as atrocious and inexcusable. 3 

Two excuses have been put forward to palliate these 
massacres. One that the beys had conspired to over- 
throw Muhammad ’Ali; the other that he had been 
instigated by the Divan of Constantinople. Both may 
be quite true. But the real reason certainly lies else- 
where. The pasha’s power was still unstable; he had 
been repeatedly called on to undertake an expedition 
into Arabia; he could not reduce his own force and 
leave the beys in a position to overthrow his power. The 
very reason which compelled Taimur to slaughter his 
prisoners before Delhi urged Muhammad ’Ali to exter- 
minate the Mamelukes. Nor is there the least reason to 
suppose that he hesitated to act thus, once he perceived 

1 Letter to Missett, May 6, 1813 (F.O. 24-4). 

2 Driault, U Empire de Mohamed Ali (1814-23), p. 113. 

3 Idem, p. 184. 



that his personal position was at stake. He was never a 
bloodthirsty man, delighting in slaughter for its own 
sake. But neither was he ever inspired by that tender- 
ness for human life which has grown up in the West in 
the last century. He recognised many reasons as 
entirely justifying the taking of life. Nor in this was 
there anything peculiar. Everyone who attended his 
divan, his friends and associates, his officers, his 
superiors, would have deemed him mad had he thought 
otherwise. In the very next year J alal-ud-din, Governor 
of Aleppo, murdered the leaders of his janissaries en 
masse . 1 Muhammad ’Ali had after all only accom- 
plished with wholesale success what the Capitan Pasha 
had attempted but bungled a few years earlier. 

From the standpoint of Turkish ethics there is little 
more to be said, and at this time Muhammad ’Ali’s 
views and outlook were essentially Turkish. Obviously 
it could not be otherwise. His birth, up-bringing, and 
experience had all tended to produce a vigorous but 
exceedingly unenlightened ruler, who would shrink 
from nothing for the attainment of his personal ends. 
The remarkable thing is not that Muhammad ’Ali set 
up his rule like a Turk, but that he was capable, as no 
other Turk of his period, of development, of absorbing 
new ideas, of adapting them to new and different 
circumstances. His keen eyes revealed to him the 
fundamental weaknesses of existing oriental rule ; and, 
alongside of the perpetual weaving of a most dexterous 
policy for the maintenance of his own position and the 
assurance of his family’s future, there went a con- 
structive power, a sense of the forces by which states 
are built up and broken down, a ceaseless struggle for 
improvement, a never-dulled consciousness of the 
defects of his administrative machinery, such as no 
oriental ruler had shown since the days of Akbar. His 
1 Barker, Syria and Egypt, i, 138-40. 


government marks a great turning-point in the history 
not of Egypt only but of the Near East as a whole, for 
he led the way in adapting western political ideas to 
eastern conditions. We must remember this as well as 
that holocaust by the Bab-al-Azab. Many years after, 
a European visitor regretted that the strange episodes 
of his rise to power should have remained so little 
known. “I do not love this period of my life”, 
Muhammad ’Ali answered; “and what would the 
world profit by the recital of this interminable tissue of 
combat and misery, cunning and bloodshed to which 
circumstances imperatively compelled me?. . .My his- 
tory shall not commence till the period when, free from 
all restraint, I could arouse this land . . . from the sleep 
of ages .” 1 

1 Puckler-Muskau, Egypt under Mehemet Ali, i, 317. 

Chapter II 


Muhammad ’Ali’s establishment as the real as well as 
the titular ruler of Egypt was followed by a period of 
some twenty years during which circumstances drove 
him generally to pose as the active, zealous and 
obedient servant of his august master the Sultan of 
Rum, the Caliph, “the Shadow of God upon Earth”. 
His obedience was indeed unreal, his zeal affected. 
From the day when the idea of seizing the government 
of Egypt first occurred to him as a practicable measure, 
he had probably always nursed the thought of ruling, 
not on behalf of another but as an independent 
sovereign. His companion-in-arms, Tahir Pasha, had 
dreamt of independent rule; his compatriot, ’Ali of 
Janina, virtually accomplished it. He himself had 
offered alliance to both the English in 1812 1 and the 
French in 1810 2 if only either of them would recognise 
him as ruler of Cairo; and he had actually proposed to 
the Divan at Constantinople in 1810 that he should be 
recognised on the same footing as the Barbary States. 3 
But both the English and the French, in view of the 
European position and existing alliances of the Sultan, 
had rejected his proposals; and he had under-estimated 
the price at which the Divan set the favour he de- 
manded. These rebuffs in no wise changed his views; 
but they did lead to their concealment for a while. 
The lack of the European alliance which he sought 

1 Missett, June 20, 1812 (F.O. 24-4). 

2 Driault, U Empire de Mohamed Ali (1814-23), p. 93. 

3 Sabry, V Empire egyptien, p. 37. 


prevented any open breach with the Porte; and 
although he seldom obeyed orders which could not be 
diverted to his own aggrandisement, his public 
language was always that of the loyal and devoted 
vassal. Throughout this period there is an ever piquant 
contrast between the professed object and the real 
purpose of his conduct. 

At this time the internal condition of the Turkish 
empire closely resembled that of the Moghuls in the 
early eighteenth century. Both were rotten to the core. 
The Divan of Constantinople, like the durbar of Delhi, 
was engrossed by the private interests of individual 
ministers. The pashas of the Turkish provinces, like 
the nawabs of the Moghul subahs, were bound by so 
slight a thread to the central government as to be 
almost independent. Baghdad and Cairo, like Hydera- 
bad and Lucknow, were almost separate capitals. But 
there was a vital difference in the political environ- 
ment of the two decaying empires. The Moghul’s 
neighbours, Marathas and Afghans, were too remote 
one from the other, and their political conduct guided 
too much by the law of nature, for either to be tempted 
to uphold Delhi for fear of the other’s getting the lion’s 
share of the spoils; whereas the Sultan’s territories 
touched a whole group of closely inter-knit European 
states, inspired by a most watchful jealousy of each 
other’s expansion. So that while the Moghul empire 
was left to dissolve by natural process into chaos, the 
Turkish dominions were held together by pressure 
from without long after they had ceased to possess any 
internal coherence. Muhammad ’Ali’s conduct as a 
pillar of the empire was dictated by this situation. 

He had small cause for gratitude. The position of the 
Divan had been consistent only in its hostility. It had 
first accused him of conspiring with the beys to his 
private benefit and the injury of the state; and when he 


sent their heads to be set up at the Great Gate of the 
Seraglio, he was reproached with having murdered the 
Sultan’s most reliable supporters. 1 

Even while he had been struggling with the Mame- 
lukes for the mastery of Egypt he had been repeatedly 
called upon by the Porte to undertake the suppression 
of the sect of Wahabis in Arabia, but until 181 1 he had 
constantly pleaded the dangers arising from “these 
miserable Egyptian chiefs”, their encouragement by 
the neighbouring pasha of Syria, or the difficulty of 
procuring shipping in the Red Sea. 2 So in acceding at 
last to the demands of Constantinople, Muhammad 
’Ali certainly was not inspired by any empty sentiment 
of obedience. He resolved to undertake an expedition 
into Arabia on the solidest of grounds. It would keep 
busy those turbulent soldiers who had fired upon him 
even when the Mamelukes were still unsubdued and 
living, and who might be yet more turbulent when no 
force remained in Egypt capable of resisting them ; and 
it would raise high his repute in the world of Islam if he 
drove the heretics from the Holy Cities. 

The Wahabis were a sect that had sprung up in 
Arabia about the middle of the eighteenth century. 
Their founder, Muhammad ibn Abdul Wahab, after 
studying in the schools of Damascus and Baghdad, had 
dwelt for a while at Mecca. There continued contem- 
plation of the lives and manners of the pilgrims 
strengthened and confirmed his rising belief that Islam 
had been corrupted from and should be restored to its 
early purity and simplicity. He therefore began to 
attack the luxury of his age and the dialectical subtle- 
ties by which deviations from the plain intention of 
Koranic texts were excused or justified. He began to 

1 To Nakib Effendi, Rajab 5 and 19, 1226 (Abdine Archives). 

2 To Nakib Effendi, Zilhaj 5, 1225, and Muharram 1, 1226 



expound these puritan doctrines in his native village in 
Najd; but having no honour in his own country, and 
being moreover at this time an unarmed prophet, he 
fled, like his prototype, and found refuge with Amir 
Muhammad ibn Saud at Darayyah. The amir adopted 
the new doctrines, and there rose in Najd a barbarous 
theocracy warring ceaselessly on the corrupt Muslims 
who surrounded it, renouncing the Turkish caliphate, 
and defying the neighbouring pashas of the empire. In 
the enfeebled condition of the state, the movement for 
some time met little effective opposition; and it was 
able to display its impartial hatred of both Shiah and 
Sunni by sacking the most sacred shrines of either sect 
at Karbela and Mecca, and Medina, and slaughtering 
devotees by the hundred within the holy precincts 

Their occupation of the Hijaz produced great com- 
motion in the Muslim world, for it completely inter- 
rupted the annual pilgrimages to the Holy Cities. In 
1805 and again in 1806 the great Syrian caravan was 
turned back. On this the Pasha of Damascus was 
disgraced and superseded. Indeed this was well 
merited. The miri, or land revenue, of the provinces of 
Damascus and Syrian Tripoli had been specifically 
assigned (according to Turkish methods of finance) for 
the cost of conducting and protecting the caravan of 
pilgrims; and the Wahabi occupation of Hijaz had 
seemed to the pasha a heaven-sent chance of allowing 
the pilgrim caravan to lapse and of diverting the miri 
into his own coffers. He had not therefore made any 
very serious efforts to disturb the Wahabis in their 
occupation of Mecca and Medina. 1 

The Sultan had for some years been issuing unavail- 
ing orders to the Pashas of both Baghdad and Damascus 
to expel the invaders from the sacred territory. Indeed 
1 Burckhardt, Nubia, p. xxxiii. 



the protection of the Holy Cities was traditionally 
regarded as a great honour, and the expulsion of the 
Turks as a great disgrace. The Porte turned therefore 
to the rising Pasha of Cairo. It would be a great stroke 
if he could be induced to exhaust his resources and use 
up his troops in destroying the Wahabis, for this would 
restore not only the Hijaz but Egypt as well to the 
effective control of the Sultan. So the Sultan and pasha 
were at last united (though for the most different 
reasons in the world) in a common desire to reconquer 
the cradle of Islam. 

Accordingly in the latter part of 181 1 the pasha’s son 
Tussun, whose earlier setting-out had been so tragically 
attended by the beys, really began his march. This 
time it was followed, instead of being preceded, by a 
tragedy. The expedition embarked at Suez and landed 
at Yam bo, but was caught in a narrow defile on the 
way up to Medina early in 1812, and the survivors fled 
back to Yambo, after three days’ fighting, with the loss 
of all their artillery. 1 The retreat had been begun by 
Tussun’s principal lieutenant, who indeed reached 
Yambo in safety, but was at once beheaded by Muham- 
mad ’Ali’s orders to promote fortitude among the 
rest. Advantage was also taken of this disaster to 
get rid of certain Albanian leaders whose turbulence 
had caused a good deal of anxiety to the pasha. 
Discredited by their defeat and disgusted with the 
hardships of Arabian campaigning, where the booty at 
best amounted to no more than a few camels as against 
strong risks of getting one’s throat cut, they were not 
unwilling to comply with the pasha’s proposal that 
they should seek more profitable service elsewhere. 

These measures, with the preparations for a new 
campaign, filled up the hot weather of 1812. The 
preparations included the seduction of Arab tribes in 
1 Missett, February 16, 1812 (F.O. 24-4). 


the Hijaz, by methods familar to us in recent times, in 
order to facilitate the advance on Medina. These 
measures succeeded. The Wahabis were driven from 
Medina in November, from Mecca and Jiddah early 
in the following year, so that the Hijaz came into 
Muhammad ’Ali’s possession, and the caliph was once 
more prayed for in the mosques of the Holy Cities. 1 

Later in the year Muhammad ’Ali in person visited 
Mecca “for the purpose of establishing order” in his 
new conquests. 2 The event showed that his plan was to 
set up a new ‘ sharif’ in Mecca, since the old one was 
believed not only to have sympathised with and 
encouraged the Wahabis but also to be possessed of a 
large treasure. The deposition was successfully accom- 
plished. The sharif and his three sons were seized and 
sent off to Cairo. 3 But this transaction alarmed a 
number of Arab tribes and the Wahabis began to 
gather once more in the desert. More men were 
summoned from Egypt to meet this reviving danger. 
The pasha ordered 10,000 to be sent with the least 
possible delay. As there were not many more than 
12,000 in Egypt, the demand could only be met with 
vigorous recruitment. Moors from the Barbary states, 
Sudanese slaves, Greeks and even Armenians were 
enlisted and gradually sent off. 4 But this new campaign 
of 1814, like Tussun’s, opened with disaster. A detach- 
ment two days’ march beyond Taifa was suddenly 
attacked by the Arabs. Ten out of the twelve chief 
officers fled at the first onset, carrying their men with 
them. Muhammad ’Ali mounted a dromedary and 
hastened to meet the fugitives, but neither threats nor 
promises could induce them to rally. As a consequence 

1 Missett, November 4, 1812 (F.O. 24-4). 

2 The same, October 14, 1813 (ibid.). 

8 The same, January 12, 1814 (F.O. 24-5). 

4 The same, April 9, 1814 (ibid.). 



seven commanders were degraded and sent back to 
Cairo; it is said that the other three were beheaded. 1 
A further defeat occurred in an attack on Taraba 
under Tussun. The force was misled by its guides, 
and its camp was rushed by the Wahabis at night. 
All the Egyptian baggage and artillery was lost. 2 The 
result was widespread discouragement. “An intelligent 
traveller” — by which term Major Missett seems to 
disguise the famous Burckhardt — who was at Jidda in 
August, observed that “in general the soldiers are 
broken-spirited, sickly, discontented on account of the 
expenses of living which are at least double what they 
are in Egypt, without any hopes of gain or plunder, 
there being here neither fellahs to fleece nor villages to 
sack. Their enemies are naked Bedouins, and a starved 
camel is the most valuable booty they can hope to 
make.” 3 

However at this moment fortune turned. In April, 
1814, ibn Saud had died and his three sons could not 
agree among themselves. 4 * Reinforcements were pro- 
cured, tribal chiefs conciliated and bribed, and the 
pasha in person took the field after celebrating the 
Bairam festival at Mecca. The Wahabis had assembled 
a great force (40,000 men, it is said) at Biselah, twelve 
hours’ march west of Taraba. Muhammad ’Ali 
attacked, and after a most bloody conflict (the descrip- 
tion is his own) the Wahabis fled, pursued for an hour 
and a half by the Egyptian cavalry. Their camp, 5000 
camels, and much baggage fell into the victor’s hands. 6 

This success ought to have led to the complete 
overthrow of the Wahabis, but for various reasons 

1 Missett, April 9, 1814 (F.O. 24-5). 

2 The same, June 6, 1814 (ibid.). 

3 Enel, of August 7 in Missett, December 7, 1814 (ibid.). 

4 Missett, July 9, 1814 (ibid.). 

6 Muhammad ’Ali to Missett, 7 Safar, 1230 (F.O. 24-6). 



failed to do so. The pasha had now been over a year 
absent from Cairo ; on at least one occasion the Porte 
had attempted to remove him from the pashaliq of 
Egypt, 1 and the return of Napoleon from Elba promised 
renewed confusion in Europe from which political 
profit might be secured. 2 He therefore entrusted the 
completion of the campaign to his son Tussun, and 
Tussun, as before, exhibited his incapacity. He began 
an advance which should have conducted him to the 
Wahabi capital of Darayyah; but found himself 
running short of stores and provisions. The Wahabis 
under their old leader would probably have inflicted a 
heavy defeat on the Egyptian invaders. But Abdullah, 
the new amir, had been severely shaken by the victory 
of Biselah. He hesitated to attack, while Tussun 
shrank from advancing. They therefore made a peace, 
by which the Wahabis renounced all rights over the 
tribes in the area occupied by Muhammad ’Ali. As 
this left the Wahabis in possession of districts north and 
east of Medina and between Medina and Mecca, 3 the 
arrangement was a mere truce to be continued until 
one side or the other should choose to renew the war. 

, Early in January, 1816, when exhausted Europe had 
relapsed into a long-unaccustomed peace, news was 
received, or at all events given out in Cairo, that some 
of the Arab tribes had broken into rebellion at Wahabi 
instigation. Tussun had fortunately died, from the zest 
with which he gave himself to the joys of Egypt after 
his desert campaigns. The command of the new expedi- 
tion was therefore entrusted to the pasha’s second son, 
Ibrahim, “the lion of the brave, whose counsel hath 
always proved fortunate”. 4 The new commander was 

1 The conspiracy of Latif Pasha. Missett, November 13, 1813 

(F.O. 24-4). 2 Cf. Burckhardt, Arabia , 1, 149. 

3 Missett, January 13, 1816 (F.O. 24-6). 

4 The same, March 8, 1816 {ibid.). 



to play a great part in the events of the coming years. 
He had been born at Kavala in 1789, and was now 
twenty-six. 1 Of low stature but powerfully built, he 
enjoyed abundant energy and could resist alike the 
fatigues of pleasure and of war. He had clear blue eyes, 
a high forehead, a fair beard. He was very active in 
both mind and body. Uneducated, like his father, he 
possessed, like his father, a rare combination of courage 
and prudence. He lacked his father’s charm of manner 
and his insight into men and situations. He was 
austere and would overawe where his father charmed. 
He would never have raised himself so high out of 
obscurity as Muhammad ’Ali had done, but in any 
event he would have been a soldier of mark, and he 
became his father’s right hand, regarding him with 
filial awe and obedience, and carrying out his orders 
with scrupulous fidelity. He possessed too his father’s 
love of looking into matters for himself, instead of 
implicitly trusting the reports which were made to 
him. 2 

His early measures were directed, not so much at 
securing any particular military success, for which he 
judged the time was not yet ripe, as at winning over to 
the Egyptian side a number of the principal chiefs, 
who had begun to weary of Wahabi control. “The 
talents ... which he has evinced”, writes Missett’s 
successor, Henry Salt, early in 1817, “in managing the 
different tribes of Bedouin Arabs give some promise of 
his ultimate success.” 3 The same agent rightly ascribed 
his success to “an undaunted firmness, or rather 
cruelty, to those who oppose him, to his having a 

1 The story was put about that he was the son of Muhammad 
’Ali’s first wife by a former husband. But this was untrue. Camp- 
bell, July 30, 1839 (F.O. 78-375). 

2 Lane-Poole, Stratford Canning, 1, 469; Campbell, Report on 
Syria (F.O. 78-283). 

3 Salt, April 28, 1817 (F.O. 78-89). 


command of money, and being celebrated for scrupu- 
lous adherence to his word — three qualifications 
peculiarly requisite to gain influence over the Arabs”. 1 
At the same time his control over his subordinates was 
a very different thing from Tussun’s loose command. 
An incident related by Salt illustrates this very 
effectively. Hasan Aga, in charge of the Hijaz frontier, 
fell into an ambush. Instead of being the first to flee, 
“the Aga shot his horse in front of the line and shared 
the fate of his men”. 2 If Ibrahim could inspire his 
followers with this noble sense of duty, he deserved to 

Abdullah ibn Saud had felt the completest security 
in his desert fastness of Darayyah. But, when all his 
plans had been laid and his preparations completed, 
Ibrahim began his advance. He moved as a friend and 
protector, not as a conqueror. Every skin of water, 
every date, every piece of wood, was promptly and 
generously paid for. His severe discipline kept his 
troops in a wholly unaccustomed way from pillage and 
insult. This ensured him something of the support 
which British armies in India had usually encountered 
on the march. But even so his expedition, based on the 
distant Red Sea port of Jidda, almost sank under the 
difficulties that beset it. No less than 80,000 camels 
were employed on his line of communications, 3 and he 
arrived before Darayyah with less than 6000 men. 
For three months he lay before the place able to 
accomplish little, and then his magazine was destroyed 
by fire. Many leaders would have spiked their guns 
and led their troops to massacre and starvation on the 
long road back. But Ibrahim held his ground, and 
repulsed the enemy’s attacks, until he received fresh 
supplies of powder and reinforcements of men. Then 

1 Salt, June 6, 1818 (F.O. 78-91). 2 Ibid . 

3 Ibid. 


he pressed the siege, and at last stormed the place, in 
September, 1818. Two shaikhs, the leaders of the 
Wahabi sect, were captured. Ibrahim shaved their 
venerable beards, plucked out their teeth, and so 
exposed them to the derision of the populace. 1 At the 
same time a large number of the ruling family was 
deported to Cairo, 2 and Abdullah ibn Saud sent on to 
make his peace if he could with the Sultan. 

For the moment the Wahabi peril had been crushed. 
The strong arm of Ibrahim, the organising energy of 
Muhammad ’Ali, had accomplished what the pashas of 
Syria and Baghdad had lamentably failed to achieve in 
spite of their relative proximity to Darayyah, and in 
spite of Ibrahim’s neglect of earlier schemes for 
persuading the sectaries by peaceful argument that 
they had wandered from the path of salvation. The 
Sophy of Persia sent Muhammad ’Ali a precious 
scimitar, its hilt and scabbard encrusted with precious 
stones. 3 Even the Sublime Porte was gratified by an 
unaccustomed degree of success which enabled it to 
behead the principal rebel, and named Ibrahim wali of 
the provinces of the Hijaz and Abyssinia, 4 while the 
English consul-general at Cairo rejoiced at the destruc- 
tion “of a band of robbers who had proved themselves 
more bigotted, intolerant, and far greater enemies to 
the progress of civilization than the very followers of 
that religion which it was their object to supplant”. 6 

The Turkish empire, like that of the Moghuls, the 
Marathas, the Persians, or even the Chinese, rejoiced 
in the most elastic of boundaries, which enabled the 
imperial government to resent or to ignore, to profit by 

1 Salt, October 8, 1818 (F.O. 78-91). 

2 Douin, L’jGgypte de 1802 a 1804, p. 46. 

3 Driault, V Empire de Mohamed Ali, p. 176. 

4 The Abdine records contain a copy of the farman, Rabi-al- 
awal 4, 1 237. 

6 Salt, ut supra . 





or to disavow, the aggressions of neighbouring rulers or 
of its own provincial governors. Beyond the provinces 
under the actual or nominal administration of the 
Sultan, there always stretched vague areas where 
Turks had once appeared as conquerors, where local 
chiefs had been frightened into a temporary submission, 
or where, in accordance with the universal ideals of 
Islam, the Caliph ought to be recognised. These vague 
claims, which could not for a moment stand the tests of 
European jurists, extended down the Red Sea and 
beyond it to Aden, and across it to the small ports like 
Massowa and Suakin on its African shore. Hence the 
inclusion in Ibrahim’s pashaliq of titular authority over 
Abyssinia. It meant in fact no more than the power of 
appointing governors at the ports in order to collect 
tolls on the produce of the Sudan — gum, ivory and 
slaves — carried down by caravan for sale to the Guja- 
rati traders frequenting the Red Sea ports. 1 

Muhammad ’ Ali, however, was far from content with 
this restrained authority. He desired to control the 
trade himself. He believed that the Sudan and 
Abyssinia abounded in gold. He knew that from the 
south came those dark-skinned stalwart slaves who had 
always been greatly prized in Egypt. All three motives 
were strong. But at the moment it would have been 
hard to tell whether he was more attracted by the idea 
of discovering mines of gold that would enable him to 
win over the whole Divan at Constantinople or the 
hope of levying and training whole armies of Sudanese, 
who would enable him to dispense with mutinous Al- 
banians and T urks and to defy the Sultan and all his hosts . 

Accordingly in 1820 preparations were made for a 
great southward expedition, ostensibly to resent some 
insult said to have been offered by the Sultan of Sennar 

1 Valentia’s observations enclosed in his letter to Canning, 
September 13, 1808 (F.O. 1-1). 


and to clear the way for the trading caravans once 
more to descend the Nile to Cairo. By the middle of the 
year 5000 troops had been assembled at Wadi Haifa, 
beyond which the pasha’s authority extended no great 
distance, and the command of the expedition was 
entrusted to Ismail, a third son of Muhammad ’Ali, 
sent to acquire experience of war and government. 1 
He succeeded speedily in overrunning the territory of 
Sennar, the eastern portion of the Sudan, subduing the 
maliks or chiefs of that area with little resistance. In 
this there was small credit, for the Sudanese were 
entirely unacquainted with the use of firearms and they 
were besides divided among themselves under two 
rivals who aspired to the chief authority. One 
murdered the other, and then fled to Abyssinia, while 
the nominal king at once submitted. The Egyptian 
army then moved farther south, ultimately reaching a 
point between the 10th and nth degree of north 
latitude. 2 But this proved much less successful than the 
earlier part of the invasion. The broken jungles were 
difficult to penetrate and were easier to defend. The 
invaders suffered from dysentery and the fevers of the 
country. Their supplies of provisions gave out. Ismail 
found himself obliged to retire again into Sennar. 

Meanwhile another body of troops under the pasha’s 
son-in-law, the Daftardar Bey, had advanced into 
Kordofan, the western half of the Sudan. After a much 
severer resistance than was met with in Sennar, A 1 
Obeid was taken and pillaged. The Sudan was thus 
conquered. But its administration was confided to 
inexperienced hands. Muhammad ’Ali had intended 
that Ibrahim should organise the province, but he was 
seized by dysentery soon after his arrival and returned 
to Egypt. The expected gold mines were not discovered. 

1 Salt, June 30, 1820 (F.O. 78-96). 

2 Deherain, Le Soudan tgyptien> p. 86. 




By March, 1822, only 500 negroes capable of bearing 
arms had reached Assouan, 1 instead of the army that 
had been expected. And Ismail himself proved a bad 
governor. Muhammad ’Ali had found it constantly 
necessary to admonish his son to employ gentler 
methods, to act with justice, to conciliate the people. 2 
But at the same time he was perpetually demanding 
fresh supplies of slaves, who could only be obtained by 
repeated raids upon a terror-stricken people. It is 
difficult simultaneously to conciliate and to enslave a 
population, and Ismail doubtless considered the 
demand for slaves the more urgent matter of the two. 
Late in 1822, returning down the river, he landed 
opposite Shendy, and demanded of the local malik 
within three days the payment of 15,000 dollars and 
the delivery of 6000 slaves. The malik declared that he 
could not do so. “Will you insult me, slave?” cried 
Ismail, and struck him across the face with a small 
Indian riding-switch. On this another malik inter- 
posed with promises of obedience, and they then retired. 
But their object was only to collect their followers. As 
soon as this was done, Ismail’s troops on one side of the 
river were pinned to their ground by a night attack 
which broke on them quite unexpectedly, while Ismail 
on the other awoke to find the house in which he was 
sleeping had been set on fire, and he and his suite were 
cut to pieces. 3 

But the unfortunate maliks had forgotten the Daftar- 
dar Bey in Kordofan. The news of Ismail’s murder 
brought him back in hot haste to Sennar, and there he 
took a most bloody vengeance. He is said to have 
massacred 30,000 persons. Other troubles sprang up. 

1 Salt, May 30, 1822 (F.O. 78-112). 

2 Letter to Ismail, Rabi-us-sani 9, 1236. 

3 Salt, December 14, 1822, and May 3, 1823 (F.O. 78-1x2 and 
" 9 )- 



A mahdi arose and obtained a considerable following. 
The European consuls were informed that he had been 
taken and beheaded, but a couple of months later he 
was still at large and reinforcements were being sent 
from Assouan to suppress him. 1 But these movements 
also were gradually crushed, and by 1826 the Sudan 
was quiet enough for Muhammad ’Ali to be taking 
measures for its development. Eight of the most 
notable village head-men of Lower Egypt, with a 
hundred and ten other persons, were to be sent to the 
Sudan to teach the natives the Egyptian manner of 
cultivation. 2 

This measure seems to have had no particular result. 
Hunger is probably the sole motive strong enough to 
make primitive races, like the Sudanese, industrious; 
and they seem to have learnt nothing from their 
Egyptian instructors, who moreover cannot be supposed 
to have been exceedingly zealous in their enforced 
task. In the next ten years the only notable change was 
the rise of Khartum from a petty village into a town of 
over 500 brick-built houses, with barracks, storehouses, 
and many gardens growing excellent figs and grapes. 
This was the work of Kurshid Pasha, who for several 
years ruled the province and made the place his 
capital. Its growth was the consequence which has 
always followed (especially in the East) the establish- 
ment at any point of the seat of government. 

But Muhammad ’Ali was most dissatisfied with the 
stationary condition of production in the province. In 
1838-9 he spent some six months inspecting it, partly 
in the never-dying hope of finding gold, but principally, 

1 Salt, April 28, 1824 (I.O., Egypt and the Red Sea, 7). Also 
letter to the General of Kordofan, Shawal 17, 1239 (Abdine 

2 Letter to the Mudirs of Lower Egypt, Rajab 11, 1241 




no doubt, in order to promote and extend the cultiva- 
tion of the land. The accounts of his tour exhibit at once 
the slight results which had been obtained and the 
ideas of development which the pasha cherished. In 
spite of or perhaps because of the high yield which 
could be obtained — 6o-fold was the current estimate — 
agriculture was still neglected, and the ground was only 
scratched, by way of cultivation, with large pieces of 
wood. Another experiment was therefore decreed in 
order to promote the growth of sugar-cane, cotton and 
indigo. A number of Arabs who had been trained at 
the School of Engineering were offered grants of ioo 
feddans of land tax-free for five years, with an assign- 
ment of a number of young Sudanese who were thus to 
be taught the more advanced methods of Egypt. At 
the same time the pasha exhorted the chiefs to seek and 
promote improvement. “If you follow the example of 
others”, he said, “you too will rise from the level of 
beasts to that of men; you will acquire wealth and 
learn to enjoy pleasures which your ignorance does not 
allow you even to imagine.” But for this labour was 
needed, and without it nothing could be secured. His 
audience is said to have been so moved by the prospects 
held out to them that they begged to be taken to Egypt 
to learn its arts; but they were told that they had better 
send their sons. 1 Since this occurred at the close of the 
active period of Muhammad ’Ali’s career, we must 
conclude that his conquest of the Sudan had for the 
time being established Egyptian authority there, had 
provided the pasha with a certain number of slaves, 
but had not affected the primitive culture of the people 
or (what was of much higher interest to Muhammad 
’Ali) their material production, just as the overthrow of 
the Wahabis had done no more than reopen Mecca and 
Medina to the pilgrims. 

1 Campbell, No. 28, May 8, 1839 (F.O. 78-373). 


In another direction, however, the expansion of 
Muhammad ’Ali’s power eastwards and southwards 
had produced important results. The French had 
remained mere spectators; but the English were more 
directly interested, and to this period, 1811-22, may be 
traced the origin of their distrust of Muhammad ’Ali’s 
policy. His military action in Arabia and the Sudan 
affected three regions in which they were already 
interested — the Red Sea, the Persian Gulf, and Abys- 
sinia. Much of the trade of those areas passed through 
the hands of Banyan merchants — mostly Gujaratis — 
sailing from Surat and other ports of western India. 
Even at the height of its power the Moghul empire had 
never been able to protect Indian ships at sea. Akbar 
had been forced to seek passes from the Portuguese; 
later emperors had obtained Dutch or English convoy. 
In the middle of the eighteenth century, six years 
before the Company obtained the diwani of Bengal, it 
had acquired the office of admiral of the empire, with 
the revenues and districts thereto annexed, and for 
seventy years afterwards the Bombay marine regularly 
convoyed Indian shipping to Basra and Jidda, flying 
the Company’s colours at the peak and the Moghul’s 
at the main. 1 The decay of Persian and Turkish power 
rendered convoy indispensable. Piracy grew rapidly, 
and was not at all discouraged by the strange leniency 
with which captured pirates were released, or the folly 
with which merchants were allowed to export timber 
from Bombay to repair and build the craft which 
preyed upon them. 2 This rising evil was encouraged 
morally, politically, and practically by the Wahabi 
movement. The Wahabis themselves formed a pirate 
fleet at Kamfuda, south of Jidda, and on occasion were 
joined by the pirates of the Gulf. 3 In 1808 a British- 

1 Low, Indian Navy , 1,151. 2 Idem, 1, 319, 336. 

3 Burckhardt, ap. Missett, March 9, 1815 (F.O. 24^6). 



owned ship was captured, and the throats of her crew 
ceremonially cut ; and in the same year the Company’s 
armed sloop Sylph was taken. 1 A punitive expedition 
destroyed a considerable number of pirate vessels in the 
Gulf; and finally a strong expedition equipped at 
Bombay in 1819 not only captured the pirate strong- 
hold of Ras-al-Khaima with the help of the Imam of 
Maskat but forced nearly all the maritime Arab tribes 
of the Gulf to enter into treaty relations with the 
Company, obliging themselves to abandon not only 
piracy but the slave trade as well. 2 In this measure the 
Company’s Government had hoped to obtain the co- 
operation of Ibrahim after his capture of Darayyah ; 
but at that time Muhammad ’Ali did not care to look 
so far afield, and the Company’s proposals came to 
nothing. 3 

Matters had followed a less troubled course in the 
Red Sea. Napoleon’s conquest of Egypt had directed 
attention thither, a flying survey had been made of it 
in 1795, and in 1804-5 Lord Valentia insisted on 
returning that way at the end of his Indian tour. He 
had two objects in view. One was the best method of 
blocking the Red Sea against any hostile advance from 
the West, the other the increase of Indian trade. He 
therefore visited all the principal ports from Aden 
onwards, and collected information about the course of 
trade. He inclined towards the occupation of Aden, and 
an alliance with the Wahabis and Abyssinia as the best 
means of securing both objects. 4 But his proposals 
remained without effect, except that Henry Salt, who 
had accompanied him on his eastern voyage and who 

1 Low, Indian Navy, 1, 320. 2 Idem, 1, 342 sqq. 

3 Sadleir’s instructions, April 13, 1819 (Sadleir, Diary, pp. 

138 sqq.). 

4 Valentia, Observations, enc. in letter Sept. 13, 1808 (F.O. 



was afterwards to become consul-general at Cairo, was 
sent on a mission to Abyssinia in 1809 in the hope of 
developing trade between that country and Bombay. 1 

At this time the East India Company was repre- 
sented by an agent at Mokha with an assistant — 
Belzoni, who played a part of some note afterwards in 
the early history of excavations in Egypt — who was 
moved about to Aden or elsewhere as circumstances 
required. At the moment, as has been already shown, 2 
Muhammad ’Ali’s principal aim was the restoration of 
the Egyptian finances, mainly by means of trade. He 
not only supplied grain to the English victuallers in the 
Mediterranean, but also made proposals to the Com- 
pany’s government in India for the development of 
trade eastwards. 3 Those suggestions were thought 
important enough for Belzoni to be sent up to Cairo, 
where a provisional treaty was arranged and signed on 
May 28, 1810. It declared that the Turkish capitula- 
tions should form the basis of intercourse with India ; 
that in the event of an Anglo-Turkish war the pasha 
would under no pretext whatever molest British 
subjects or property, but on the contrary afford them 
every protection; that deserters from English vessels 
should be restored, even if they had embraced 
Islam; 4 that travellers with their personal baggage 
should pass free of duty; that caravans of merchandise 
should be escorted to and from Suez at a charge of 3 
Spanish dollars the camel-load ; and that customs-dues 
should be levied at the ad valorem rate of 3 per cent. 5 

This agreement was never ratified, no doubt for fear 
of impairing British relations with Constantinople. 

1 Salt, Report, March 4, 1811 (F.O. 1-1). 

2 See p. 29 supra. 

3 Anni, September 28, 1808 (F.O. 24-3). 

4 A provision hitherto invariably refused by the Turks. Cf. 
Abbott, Under the Turk , p. 29. 

6 Briggs, May 30, 1810, and enclosure (F.O. 24-3). 


About the same time the English government refused to 
allow the pasha’s corvette, the Africa, to proceed to the 
Red Sea by way of the Cape . 1 For the time therefore 
the pasha hesitated, uncertain how to invest the 
alliance he desired with sufficient attraction to induce 
assent on the part of the English. At one time we find 
him prohibiting (in ironical obedience to the Sultan’s 
standing orders) Bombay vessels from proceeding 
further north than Jidda . 2 But he ultimately resolved 
upon embarking personally in the Indian trade, 
appointing Forbes and Company his agents at Bombay, 
and despatching thither a considerable quantity of 
European goods as well as a million dollars in specie . 3 
At the same time he urged again upon the English 
consul, in view of the growing depredations of the 
Wahabi pirates, the need of having “some sort of 
naval force there to repel their insults, since otherwise 
it would be no longer safe for even his sons to pass to 
and from the Hedjaz ”. The consul, Salt, now supported 
this request. “It undoubtedly would be far better”, he 
wrote, “that His Highness should have a prepon- 
derating influence there than that such pirates as the 
Wahabees should have possession of the sea. With 
regard to Egypt the pasha has become so complete a 
merchant that he has placed himself entirely at our 
merc\-, his revenue now so vitally depending upon 
commerce . . . that he could not support his government 
many months without it. The admiral commanding in 
the Mediterranean might in my opinion at any time 
bring him to our own terns, in the event of a rupture, 
without any additional force than that always under 
his command, by simply anchoring at Aboukir and 

1 Missett, February 16, 1812 (F.O. 24 -4). 

4 The same, June 6 and September 7, 1815 (F.O. 24 -6). 

3 The trade proved very disappointing and was soon abandoned. 
Salt, April 28, 1817 (F.O. 78-89). 



blockading the coast. The same thing might be done in 
the Red Sea, as two frigates stationed between Jedda 
and Suez would cut off all their communication by sea 
and soon reduce him to terms .” 1 

As a result of these arguments, the objections to the 
pasha’s sending his corvette round to the Red Sea were 
withdrawn . 2 

So far then English relations with the pasha ever 
since his rise to effective government had been decidedly 
amicable. Nor, as we have just seen, were they affected 
by his progress against the Wahabis, although indivi- 
dual Englishmen may have been disposed to back the 
latter . 3 Missett had indeed regretted Muhammad 
’Ali’s success in Arabia, but only because he feared the 
pasha would be lured on to destruction, “convinced as 
I am that were he to meet with a premature death this 
country [Egypt] would again relapse into that state of 
revolution from which he drew it ”; 4 and Captain 
Sadleir had been sent to congratulate Ibrahim on his 
success at Daravyah and propose joint action in the 
Gulf . 5 In like manner, when Salt feared that the 
expedition into the Sudan might be designed to 
embrace also the conquest of Abyssinia, and informed 
Muhammad ’Ali that such an event would be un- 
welcome in England, the pasha at once declared 
emphatically that though the country were full of gold 
and jewels, and its conquest certain, he would relinquish 
it rather than compromise his relations with Great 
Britain. “ I have never known him give his word to 
us”, Salt adds, “on a point which he did not mean to 
keep .” 6 

1 Salt, June 15, 1816 (F.O. 24-6^. 

2 To Salt, May 3, 1817 (F.O. 78-89}. 

3 Memorandum by R. Dundas, January 3, 1809 (F.O. i-t). 

4 Missett, March 9, 1815 (F.O. 24-6). 

* Vide p. 56 , supra. 

* Salt, November 20, 1820 (F.O. 78-96). 


But such relations in the eyes of the Divan were 
fraught with danger. They feared that sooner or later 
this over-strong pasha would win the English alliance 
which he sought and shake himself entirely free from 
control. They therefore seized every occasion that 
offered of stirring up trouble. They had, for instance, 
endeavoured to commit Muhammad ’Ali to a support 
of the pirates of the Persian Gulf— proteges worthy of 
their protectors. More serious difficulties arose out of 
the conduct of the governor of Mokha. In 1817 an 
Arab was detained for a short time at the British 
factory. He was released at the governor’s request, but 
the small factory guard, the captain of a merchant 
vessel by chance there, and the resident were seized, 
beaten, and abused, while the factory itself was 
plundered. After a considerable delay spent in ascer- 
taining and deliberating upon the facts, it was resolved 
to send a force to obtain satisfaction. 

The dependence of Mokha upon the Turkish empire 
was peculiarly tenuous. It was the chief port of the 
Imam of San’a, over whom the Sultan exercised neither 
influence nor authority. But in the course of 1818 
Muhammad ’Ali had handed over to him certain 
districts round the northern port of Hodeida, on con- 
dition of his supplying annually a certain quantity of 
coffee for the use of the Sultan. This was reckoned as 
tribute, and the country was thereafter “considered as 
in some degree under the protection of the Porte”. 1 
Such views however could scarcely be accepted by 
European powers which did not recognise rights un- 
accompanied by effective control. Accordingly the 
East India Company demanded reparation from the 
Imam of San’a. He followed the usual policy of delay. 
Mokha was therefore bombarded; the forts were 

1 Salt, November 19, 1820 (I.O., Egypt and the Red 
Sea, 7). 



attacked and destroyed ; 1 and the imam then yielded to 
force what he should have conceded to the former 
unarmed demands. A treaty was made by which the 
British resident was to have a guard as at Baghdad or 
Basra, and to be allowed to appear in public on horse- 
back; a cemetery was to be allotted for the burial of 
Christians; the Surat merchants were admitted to be 
under British protection ; and the customs dues payable 
by British traders lowered to the rate paid by theFrench . 2 

Thus fell a stronghold of Islam in which till then 
Christians had been exposed to unpunished insult, had 
been condemned to walk, had been forbidden to pass a 
certain gate, and had been obliged to see the corpses of 
their countrymen thrown out to be eaten by dogs and 
jackals, and where Indian merchants had been coerced 
into payment of large sums by being half-stifled with 
the smoke of brimstone . 3 Such an odious change as was 
involved in the new treaty produced a flood of sapient 
rumour. A chain-cable happened to be landed for the 
use of one of the Company’s cruisers. The story spread 
abroad that each link bore magic characters, and that it 
was to be used either to draw the whole city into the 
sea or to pluck up mountains so as to open a passage to 
San’a itself . 4 At Constantinople whither these stories 
spread, strong remonstrances were addressed to the 
British ambassador, while Muhammad ’Ali was 
severely reproached for his neglect and enjoined to 
occupy all the Red Sea ports as far as Aden in the name 
of the Sultan . 5 

The defeat of the Wahabis and the conquest of the 

1 Bruce to Salt, January 20, 1821 (I.O., Egypt and the Red 
Sea, 7). 

3 The treaty was signed Jan. 15, 1821. A copy occurs loc. cit. 

3 Salt to Strangford, August 16, 1822 (I.O., Egypt and the Red 
Sea, 7). 

4 Hutchinson to Bombay, January 25, 1823 (ibid.). 

6 Salt to Strangford, August 16, 1822 (ibid.). 



Sudan were immediately succeeded by a remarkable 
reorganisation of Muhammad ’Ali’s military forces. 
The troops by whose aid he had climbed to power, 
were hardly more than an armed rabble, impatient of 
discipline, to be kept in check only by regular pay and 
ferocious punishment, and as great an obstacle to the 
pasha’s maintenance of his position as they had been an 
indispensable aid to its creation. In 1816, for instance, 
Missett reported that a considerable part of the army 
had been cantoned along the coast, and that on his 
enquiring the reason for this measure Muhammad ’Ali 
had explained “that finding himself unable to repress 
the excesses of which for the last few months his troops 
have not ceased to render themselves guilty, he had had 
recourse to the expedient of ordering them out of the 
city ... in the hope that, when divided into small 
bodies they would be more easily brought back to 
discipline and subordination”. 1 

For these reasons Muhammad ’Ali resolved upon 
establishing a Nizam jadid — a new-model army — with 
a European mode of organisation, discipline and 
control. His retention of power would certainly 
depend upon his success in this project; and yet it was 
undoubtedly a measure of the greatest difficulty. 
Quite recently Sultan Salim had been deposed and 
murdered for having dared to introduce the manners 
of infidels into Islam by attempting to enrol his 
janissaries in new corps. The pasha however was not 
the man to turn aside from an enterprise merely 
because it was difficult and dangerous. He did not 
believe either that military reform was unwelcome to 
the rank and file but only to the chiefs who could not 
endure the detection of their long-established frauds 
upon the public treasury. 2 The justice of this view was 

1 Missett, March 8, 1816 (F.O. 24-6). 

2 Burckhardt, Arabia , 1, 146. 


immediately demonstrated upon his first attempt to 
introduce the European mode of military exercise. 
This took place after his return from the Hijaz. He 
began with the bodies under the command of his own 
immediate relatives; but when he extended it to the 
troops over whom he did not possess the same degree 
of influence, discontent at once began to show itself. 
The pasha proclaimed that any who were indisposed to 
obey orders might draw any arrears of pay due to them 
and depart from the country. But no one took ad- 
vantage of this offer, and then one afternoon a party 
of soldiers, assembled in the Usbekiah Square at 
Cairo, before the pasha’s palace, began suddenly to 
plunder the shops with cries that there was no God but 
God. The next day the mutiny spread. Many shops 
and warehouses were plundered. The Frank quarter 
was several times assaulted, and Europeans could not 
venture out except in Turkish dress. 1 For the moment 
therefore the new plans were dropped. 

But opposition, instead of weakening Muhammad 
’Ali’s purpose, set him upon devising other ways of 
carrying it into effect. As has been observed, one great 
object of the Sudan expedition was to secure supplies of 
slaves, who could be trained to war in whatever way he 
pleased. Hence the urgency with which Ismail was 
ordered to collect and send Sudanese slaves to Assouan. 
And since such persons could not be expected to 
provide material for officers, a body of over 300 young 
Mameluke slaves, Muhammad ’Ali’s personal property, 
were sent up to Assouan to be trained as well. 

The charge of this new military school was confided 
to a French officer, Colonel Seve. He had fought his 
way up from the ranks, gained a well-deserved Cross of 
the Legion of Honour, and after Waterloo retired with 
the rank of Captain. In 1819 he had come to Egypt, 

1 Missett, August 24, 1815 (F.O. 24-6). 



where he was completely conquered by the charm and 
character of the pasha. He renounced his faith, but in 
him the act betokened none of that lightness of 
character which we associate with the renegade. He 
became and remained one of the most attached and 
trustworthy servants of Muhammad ’Ali. Twenty years 
later, when Great Britain was bent on restoring Syria 
to the Sultan’s tyranny, great offers were made to 
Sulaiman Pasha (as Seve was then called) to bribe him 
into desertion. But neither the offer of a hereditary 
pashaliq nor the consciousness of a failing cause could 
move him. He answered that he was bound not only 
by the duty of gratitude but also by an unlimited 
devotion to his master . 1 

The early organisation of the Nizam jadid was 
probably the most difficult task that Seve was ever 
called upon to undertake. Military discipline adminis- 
tered by a European was so strange and so unnatural 
a thing in Egypt that at first Seve’s life was in frequent 
danger. He would exercise a party in musketry, and 
hear shots whistle by his head . 2 On one occasion he is 
said to have met a conspiracy among the Mamelukes by 
assembling the conspirators, informing them of their 
designs, and offering to fight with his sabre single- 
handed as many as chose to attack him . 3 

At first the training camp at Assouan comprised only 
the young Mamelukes and the Sudanese slaves. The 
latter proved disappointing. They were strong and 
docile enough, submitted patiently to military disci- 
pline and learnt their drill. But they refused to be kept 
alive. Trivial ailments, which would hardly have laid 
up a European or an Arab, in them ended fatally. 
They died like sheep with the rot. Some 20,000 were 

1 Medem to Nesselrode, September 26-October 8, 1840. 

2 Douin, line Mission militaire , p. xiii. 

3 Salt, February 8, 1824 (F.O. 78-126). 


thought to have been collected and sent up to Assouan 
by 1824, but in that year not 3000 remained alive. 1 
This failure, which contrasts so pointedly with our 
own experience, was perhaps mainly due to the fact 
that Muhammad ’Ali’s recruits were not free men but 

The failure of this expected military resource led, on 
the advice of Drovetti, the French consul-general, to 
the application of a similar method of recruitment to 
the fellah population of Egypt. The idea was perhaps 
originally suggested by the great success which had 
attended the application of European discipline to the 
sepoys of India. But any comparison is impaired by the 
fact that till now no one had dreamt of employing the 
despised fellah as a soldier, whereas in India the sepoy 
had practically always belonged to a military caste. 
The proposal was warmly taken up ; its unusual 
character led to a rebellion in various districts, 2 but 
about 30,000 fellahin were taken and sent up to 
Assouan, and Seve was allowed to increase his cadre of 
European instructors. These received a very bad 
character from the French officer who later was placed 
over them. He describes them as Spanish, Piedmon- 
tese, and Neapolitan refugees, void of truth, faith or 
honour; in short the worst set of rascals to be found 
anywhere in the world. 3 But under Seve at all events 
they seem to have done their work well. Salt, who 
visited the training camp with Muhammad ’Aliin 1824, 
thought the pasha had reason to be delighted with and 
proud of his new army — an opinion amply borne out by 
its coming service under Ibrahim in the Morea and in 
Syria. The most noticeable deficiency was the lack of 
an organised medical service. A school of medicine 

1 Salt, February 8, 1824 (F.O. 78-126). 

2 Driault, U Expedition de Crete et de la Moree , p. 13. 

3 Douin, Une Mission militaire , p. 22. 





could not (as Salt said) be planted like a melon-bed, 
and Arabs could be made into soldiers quicker than 
into hakims. 

In another direction also a very significant beginning 
had been made. Salt had called repeated attention to 
the manner in which the pasha’s power and commerce 
was limited and overshadowed by British naval power, 
which both in the Mediterranean and the Red Sea 
could place him under irresistible pressure. Muham- 
mad ’Ali was as acutely aware of this fact as was the 
British consul-general. It was and remained through- 
out his active career the chief reason for his consistently 
seeking to secure an alliance with Great Britain in 
preference to any other power. But he was also intelli- 
gent enough to draw from it very valid conclusions as 
to the advantages of sea power itself. It is in the last 
degree unlikely that he ever contemplated any attempt 
to rival Great Britain on the seas; but he did seek to 
secure a naval preponderance over every state with 
which he expected to be drawn into conflict. He seems 
indeed to have been the only eastern ruler who 
recognised the importance of sea power and who set 
himself deliberately to cultivate it. 

These views first unmistakably exhibited themselves 
in the course of his Arabian campaigns. The Wahabi 
pirates had threatened his sea communications between 
Suez and Jidda, and he had therefore been anxious to 
send his armed corvette the Africa round into the Red 
Sea. When he was disappointed of this by the English 
refusal of a pass, he ordered a frigate to be built for him 
at Bombay ; 1 he sought to induce a notable Arab 
pirate chief to take service with him ; 2 and he actually 
built a ship of war mounting sixteen guns at Suez . 3 

1 Burckhardt, Nubia, p. xciii. 

2 Burckhardt, Arabia, I, 282 n. 

3 Enc. ap. Missett, March 9, 1815 (F.O. 24-6). 


Altogether he must have assembled in the Red Sea a 
flotilla more than capable of holding its own against 
the Wahabis. 

A little later he applied the same ideas to the Mediter- 
ranean. He began to purchase such vessels as could be 
picked up in the Levant or built at Genoa or Venice. 
He then sought to strengthen himself by obtaining 
vessels of a superior type. In 1821 he applied to both 
France and Great Britain, desiring each country to 
supply him with a couple of frigates of the latest 
model. 1 Both powers declined. Canning wrote It is 
entirely out of the power of His Majesty’s Government 
to comply with this request, inasmuch as it would be a 
direct violation of the neutrality which the King has 
declared it to be his intention to observe during the 
present unhappy contest between the Ottoman Porte 
and the Greeks”. 2 Muhammad ’Ali then sought (and 
obtained) permission to build two frigates and a brig- 
of-war at Marseilles. He was therefore beginning to 
provide himself not only with an army framed and 
trained like those of Europe, but also with vessels 
which would enable him to meet the Greeks, and 
perhaps one day, which he could not too clearly fore- 
see, the fleet of the Sultan. 

1 Salt, November 6, 1821 (F.O. 78-103). 

2 To Salt, January 31, 1822 (F.O. 78-112). 


Chapter III 



The successful invasions of Arabia and the Sudan had 
thus led to the reorganisation of Muhammad ’Ali’s 
army, to the foundation of a naval force, and to a con- 
siderable extension of the pasha’s authority. But so far 
his progress had not brought him into collision with any 
European power. French policy at the moment was 
far from aggressive; and though individual English- 
men might view the employment of French officers 
with a jealous eye, London exhibited no trace of 
anxiety, while Calcutta was much more disposed to 
co-operate with than to resist the establishment of 
public order in important areas of Indian foreign 
trade. 1 The endeavours of the Sublime Porte to em- 
broil Muhammad ’Ali with England had so far failed. 

Then in April, 1821, the Greeks in the Morea, 
taking advantage of the rebellion of ’Ali Pasha at 
Janina, themselves broke into rebellion. There were 
some 20,000 Muslims scattered through the country 
districts. They were suddenly attacked. Those who 
could escaped into the Turkish garrisons. The remainder 
were massacred. The garrisons were then beleaguered. 
Some surrendered under promises of safety, others 
surrendered at discretion. It made no difference. All 
alike were put to death. Near Tripolitza 3000 Greeks 
defeated 5000 Turks. The result was the surrender of 

1 M. Sabry’s view ( UEmpire egyptien sous Mohamed Aly ) that 
England was hostile from the first seems based on ignorance of, 
or a failure to understand, the documents he cites. 



that place and Navarino. At neither did the insurgents 
carry out the terms of capitulation, and at Tripolitza 
they slaughtered 8000 Muslim men, women and 
children. As a matter of course these events were 
followed by like massacres of Greeks at Constantinople 
and elsewhere. The Greek patriarch and four of his 
bishops were hanged, and it was believed that at least 
one Greek life was taken for every Muslim that had 
perished in the Morea. The Shaikh-ul-Islam — the head 
of the theologians of Constantinople — was removed 
from his post for his unworthy conduct in seeking to 
stay the course of this revenge. 1 

The movement spread naturally to the islands of the 
sea. The small shipping that carried a great part of the 
Levantine trade was owned by island Greeks and 
manned by a considerable number of Greek sailors. 
A war fleet was formed, and developed the use of fire- 
ships. These tactics filled Turkish sailors with horror 
and alarm. Control of the sea meant power for the 
insurgents ashore. A national government was formed ; 
a popular assembly was convened. The Sultan might 
exact blood for blood in Smyrna and Constantinople, 
but he could not recover his lost territory. He was as 
impotent against infidel Greeks as he had been against 
heretic Wahabis. 

Muhammad ’Ali seems to have observed these events 
with benevolent detachment. In true opportunist vein, 
he had got rid of Albanian troops whom he did not 
want by encouraging them to leave his service for that 
of the rebel ofjanina. He was informed of the activities 
of the Greek revolutionary societies which had been 
established at Cairo and Alexandria, but took no 
measures against them. He did nothing to hinder the 
embarkation of Greek volunteers at Alexandria. He 
even set free a party of Greek slaves sent to him by the 
1 Driault, UExpidition de Crete et de la Morie, p. 5. 


Dey of Algiers. 1 In 1822, however, the Sultan offered 
him Crete if he would reduce it to submission. The 
island had been the scene of reciprocal massacre. Hasan 
Pasha, who had married one of Muhammad ’Ali’s 
daughters, and on his death Husain Bey, were accord- 
ingly sent. The Cretan rebels were numerous and bold, 
but were reduced by measures of undeviating severity. 
This process occupied nearly two years; but in 1824 
Muhammad ’Ali was able to report that their last 
refuge, Sphakia, had been cleared, and that the rebel 
chiefs had been executed ; in proof of his assertion he 
sent a bag of ears to be nailed to the Great Gate of the 
Seraglio. 2 Even better evidence of Husain Bey’s success 
was however to be found in the fact that he was able to 
extend his operations. Not far to the north-east of Crete 
lie the islands of Kasos and Scarpanto. The former was 
the home of numerous mariners who had enthusiastically 
supported the cause of Greek independence by preying 
upon Turkish trade. Husain Bey equipped an expedi- 
tion against these islands. The Kasiotes rejected his 
summons to surrender. Their stronghold was carried by 
storm; and the troops were then given twenty-four 
hours’ plunder, in which about 100 men were killed, 
900 women and children taken prisoners, and great 
quantities of coffee, silk and other goods — the results of 
the islanders’ privateering — made booty. As a further 
punishment 500 men were picked to serve at the usual 
rate of pay on the Egyptian vessels. The people of 
Scarpanto, on the contrary, surrendered on the first 
summons, and were only required to pay the three 
years’ tribute that they owed to the Turkish govern- 
ment. 3 The episode very fairly displays the policy of 
Muhammad ’Ali. Obstinate rebels should be relent- 

1 Politis, UHelttnisme et l' Iigypie moderne, 1, 187 sqq. 

2 To Nakib Effendi, Shaban 19, 1239 (Abdine Archives). 

8 Salt, June 24, 1824 (F.O. 78-126). 



lessly destroyed but others admitted to favourable 
terms, so as to keep equally alive the sentiments of hope 
and fear. 

Success in pacifying Crete led of course to further 
demands on the pasha. Very early in 1824 the Sultan 
Mahmud II issued a farman graciously bestowing the 
pashaliq of Morea upon him. It is unlikely that this 
damnosa haereditas was accepted out of fear of the Sultan’s 
displeasure. But there was the Nizam jadid, Which had 
done so well in Crete, to be tried out on a larger scale. 
Great Britain was still neutral, and the factors which 
were shortly to transform her into an active participant 
were still beyond the range of even the keenest eye at 
Cairo. Above all, there was the thought that the con- 
quest of the infidel following on the conquest of the 
heretic would raise the conqueror’s name high in the 
world of Islam, wholly obliterating the ill-effects of 
eating Christian-wise with a knife and fork in private, 
of drinking Christian liquor, and of protecting with a 
firm hand the lives and goods of Christians within his 
government. The subjugation of the Greeks would mark 
him out as the leader of the age, enable him if he 
pleased to defy the orders of the Sultan, and (he fancied) 
entitle him to the respect, the friendship, even the 
alliance of one or other of the great powers. 

Six months were spent in preparing the expedition. 
On July 10 it set sail from Alexandria. It consisted of at 
least 16,000 soldiers, 100 transports, and 63 armed 
vessels. 1 The command was entrusted to Ibrahim, but 
it was not as complete as his father had desired. He was 
named Wali of the Morea, with authority over all the 
troops and over certain vessels; 2 but with its usual and 

1 A letter to the Grand Vazir, Zilkaidah, 1239 (Abdine 
Archives), states that the expedition comprised 30,000 men and 
196 vessels. 

2 Letter of Shaban 15, 1239 (ibid.). 


indeed justified, but fatal, habit of dividing authority, 
the Porte gave the chief naval command — the office of 
Capitan Pasha — to Khusrau Pasha. Even in the West, 
it has almost invariably happened that when the com- 
mand of the army has been given to one man and of the 
fleet to another, the commanders have been too busy 
quarrelling to defeat the enemy. In the present case 
that result was completely secured by the selection of 
Khusrau as Capitan Pasha. He and Muhammad ’Ali 
had been declared enemies ever since the former had 
been so unceremoniously driven from the pashaliq of 
Cairo. So that while the Sultan might rest assured that 
his general and admiral would never unite to overthrow 
their master’s power, he had every reason also to expect 
that they would never lay at his feet the trophies of a 
combined victory. 

The due consequences followed. The plan of cam- 
paign had been for the Turkish fleet and the Egyptian 
expedition to rendezvous off Rhodes, to capture the 
island homes of the Greek privateers, and then to re- 
conquer the Morea. The plan was Muhammad ’Ali’s 
and displays his acute sense of the importance of the 
command of the sea. Khusrau had begun well enough. 
On July 3 he had captured Psara, a nest of corsairs just 
west of Chios. Samos was to be the next place of attack; 
but Khusrau spent a month celebrating his victory, 
with the result that off Samos he encountered a squad- 
ron of Greek vessels. In the action that followed on 
August 1 6 he lost two frigates and a corvette, and the 
Turkish squadron fled, “its sails filled with terror”. 
Meanwhile Ibrahim had arrived off Rhodes on August 
13, and on the 29th he joined the Capitan Pasha off 
Budrun, near the site of the ancient Halicarnassus. In 
September a number of conflicts followed with the 
Greeks, who invariably took the offensive, and in- 
variably had the best of the engagement, while the 



Turkish portion of the Muslim fleet displayed the 
greatest anxiety to avoid a close encounter. At the end 
of the month Khusrau was temporarily recalled to 
Constantinople. Ibrahim, left alone, stood naturally on 
the defensive, but by the close of the year managed to 
concentrate his ships and men at Suda Bay, on the 
north-west of Crete, without having undergone any 
considerable loss. In face of the hastily improvised 
character of his squadron, this negative result must be 
considered highly creditable. The Egyptian squadron, 
strong in the resolute character of its commander, had 
not yielded to the panic that Khusrau had failed to 
overcome ; and Muhammad ’Ali in Egypt was the last 
man living to give way to defeatism. “I know very 
well ”, he said, “ that I can’t form a navy in the sands of 
the Pyramids, and that I can’t help incurring losses. 
But in the long run I shall have a navy, and then I shall 
be able to meet and overcome the Greeks .” 1 

In this admirable constancy of purpose, the pasha 
busied himself in enlarging his naval forces. Four 
vessels which he had ordered from Italian shipyards 
arrived. Five more were bought (indirectly) from the 
Greek rebels. He sent a French officer back to France 
to obtain leave to build two frigates and a brig in the 
king’s dockyard under the superintendence of French 
officials . 2 These vessels were accordingly begun at 
Marseilles . 3 Greek merchants were found to lay down 
other vessels on account of Muhammad ’Ali, although 
their parents had been massacred at Chios and they 
themselves incurred by their conduct the greater ex- 
communication . 4 Other vessels were built at Venice 
and Leghorn . 5 

Meanwhile the Greek fleet had been compelled to 

1 Douin, Une Mission militaire, p. 7. 2 Idem, pp. 25-6. 

3 Douin, Les Premieres Frigates de Mohamed Aly, p. 28. 

* Idem, p. 31. 6 Idem, p. 65. 



quit their watch over the Egyptians, by the demands of 
the sailors for their pay. Ibrahim was thus enabled in 
January, 1825, to cross without opposition from Suda 
to Modon, in the south-west corner of the Morea. From 
the first the Greeks were no match for him in the field. 
A considerable body of them was defeated before 
Navarino and that place surrendered on May 18, and 
next month he was able to occupy Tripolitza in the 
centre of the peninsula. This event was followed by a 
guerilla war in which the Greeks had a better chance of 
success. However this was met by the burning of the 
villages concerned, the destruction of their crops, and 
the driving off of their cattle; so that even here the 
Greeks began to weary of the war. 

Nor did they seem able to use their superiority at sea. 
The one notable attempt which they made was a raid 
on Alexandria where they tried to burn the Egyptian 
ships lying in the harbour. In the afternoon of August 
10 a vessel ran in under Russian colours, and as she was 
nearing one of the men-of-war, she was fired, and the 
crew dropped into a boat astern and rowed off in safety 
to another vessel near the entrance of the port. The 
attempt wholly miscarried. The sails of the fireship 
taking fire, she fell off her course and drifted safely away 
from the anchored vessels. Muhammad ’Ali chanced 
to be sitting in the Ras-al-tin Palace, overlooking the 
harbour. He at once jumped upon his mule and rode 
to the battery at the point, hoping to be able to catch 
the enemy before they got out of range. Failing in this, 
he ordered several vessels to proceed in chase of the 
Greeks, and one which was unlucky enough to be ready 
for sea was obliged to sail at once alone. Next day three 
more sailed. On the 12th news came that the Greeks 
had burnt a boat bringing wood from Sataliah, within 
sight of the ship that had put out on the 10th. This was 
too much for the pasha. In unbearable anger, he 



hurried aboard a corvette on the first boat he could find 
at the water-side and immediately put to sea, where he 
remained for a week, seeking in vain both the Greek 
vessels and his own. 

Had he met the Greeks, he would almost certainly 
have perished. But he nearly fell into another even 
greater danger. The day after his departure all Alex- 
andria was alarmed at seeing a fleet of forty sail, which 
were thought to be the Greeks coming to renew their 
attack in full force. They actually proved to be the 
squadron and transports of the Capitan Pasha, forced 
by lack of provisions and ammunition to abandon the 
besiegers of Missolonghi, whose operations he had been 
covering from the sea. It is doubtful whether his arrival 
at Alexandria much reduced the anxiety of either the 
populace or the ministers. A hurried council was held; 
the French and English consuls-general were consulted; 
and it was decided to admit the Turkish fleet into the 
harbour, but under no circumstances to allow the 
Capitan Pasha to land. The story reached Cairo that he 
had detached seven of his vessels to block the entry into 
Rosetta and Damietta, and that he was resolved to 
seize Muhammad ’Ali if he could . 1 Both French and 
English consuls-general were aghast at the recklessness 
with which the latter had put to sea in a single vessel 
when most of his troops and his ablest commanders 
were in the Morea. Great was the relief when on 
August 20 it was learnt that he had entered the harbour 
under cover of darkness and had reached the Ras-al-tin 
Palace before anybody was aware of his return. 

Whatever had been Khusrau’s designs on finding his 
ancient enemy absent, they were instantly covered by 
bland congratulations on his return and polite requests 
in the name of the Sublime Porte for assistance in 
money and munitions. He hastened to be the first of 
1 Douin, Une Mission militaire , p. 6i. 


the two to pay the other a visit of compliment. He was 
met at the water’s edge by Muhammad ’Ali. They went 
together to the palace. Reaching the hall of audience 
each used gentle violence to place the other in the seat 
of honour. Each tried to seize the fly-swish to keep flies 
from settling on the other’s revered face. Supplies 
were ordered for the fleet, and Khusrau was given 
80,000 dollars with which to pay his people. 1 When at 
last in October he went away, the two pashas parted 
like brothers. With him went Muhammad ’Ali’s new 
vessels, and a considerable body of men — 1500 horse 
and 8000 infantry — intended to enable Ibrahim to hold 
his position in the Morea and at the same time to co- 
operate in the siege of Missolonghi which the Turks had 
been vainly attacking for the last six months. 2 The 
measures proved successful. Leaving Colonel Seve to 
command the Morea, Ibrahim moved across to Misso- 
longhi, and his aid enabled the Turks to storm it early 
in 1826. This was followed by the siege and capture of 
Athens. The Greek cause was visibly sinking. They had 
routed the Turks; but they were being crushed by 
Ibrahim and the regular troops that his father had 
organised and the naval force that his father had 
gathered together. 

These successes first in Arabia and then in Greece had 
something of an intoxicating influence. For the moment 
perhaps it seemed to the pasha that no limit need be 
set to the expansion of his power. He dreamt of raising 
his disciplined forces to 100,000 men. As soon as the 
Morea had been completely subjugated, he saw him- 
self handing over this uninviting territory “to its legiti- 
mate master, the Grand Signor”, recalling his troops, 
filling up the gaps in their ranks, conquering Yemen, 
holding the Red Sea and the Persian Gulf, and occupy- 

1 Salt, September 15, 1825 (F.O. 78-125). 

2 The same, October 22, 1825 {ibid.). 



ing the pashaliqs of Acre and Damascus. Then, 
Alnashar-wise, having reorganised those unfortunate 
provinces, with redoubled power he would move on the 
Tigris and Euphrates, and there consider what further 
conquests would be most profitable. “The sword”, he 
said, “has delivered power into my hands, and I should 
be indeed ungrateful if I did not continue to employ it 
for the service and salvation of the Turkish empire.” — 
“But”, said the French officer to whom these con- 
fidences were addressed, “will the English give you 
time to accomplish these great designs?” 1 

The truth was that they could not be carried into 
effect at all unless the pasha could reach an under- 
standing with Great Britain and obtain English co- 
operation. Probably he himself was as fully convinced 
of this as anyone in the world ; and the time was indeed 
approaching when he was perhaps nearer to securing 
his object than at any other point in his career. Two 
conditions had to be fulfilled in order to make a treaty 
seem worth while in English eyes. The first was that our 
relations with the Sultan must be severely strained, if 
not absolutely broken, before the separate political 
existence of Egypt could possibly be recognised. The 
second was that the pasha must have some clear advan- 
tage to offer or withhold, commensurate with the obli- 
gations implicit in an alliance. One attempt had 
already been made by way of a treaty with the Com- 
pany’s government in India; but in 1810 the develop- 
ment of trade along the Suez route had seemed of 
dubious advantage to the English authorities, and the 
treaty had never been ratified. But Ibrahim’s conquest 
of the Morea had perhaps provided a pledge of greater 

The Greek revolt had excited great interest in 
Europe. Poets and liberals hailed it in prose and verse 
1 Douin, Une Mission militaire , pp. 79-80. 


7 8 

as the rebirth of liberty. Even musty scholars in their 
closets were stirred by the fancied repetition of Mara- 
thon and Salamis. When it began to seem likely that 
the rebellion would be drowned in blood, indignation 
ran high. All the misdeeds of the Turks were multiplied 
and exaggerated. Every hint that the Greeks were a 
trifle less heroic than the heroes of classical literature 
was angrily rejected. Stories spread that Ibrahim con- 
templated the wholesale enslavement and removal of 
the Greek population and the resettlement of the Morea 
by Turks or Arabs. Even George Canning, indisposed 
as he was to romantic policies, thought the situation 
demanded intervention. “The selling into slavery”, he 
wrote to his cousin, the ambassador at Constantinople, 
“the forced conversions, the dispeopling of Christen- 
dom, the recruiting from the countries of Islam, the 
erection in short of a new puissance barbaresque — these 
are facts . . . new in themselves, new in their principle, 
new and strange and hitherto inconceivable in their 
consequences, which I do think may be made the 
foundation of a new mode of speaking, if not acting, . . . 
and one which, I confess, I like the better because it has 
nothing to do with Epaminondas nor (with reverence 
be it spoken) with St Paul .” 1 

The agrarian character of the war in the Morea and 
the traditional practice of all Muslim armies had indeed 
produced something like the conduct that Canning so 
indignantly described. As we ourselves had found, and 
were again to find, in Ireland, it was impossible to dis- 
criminate between the peasant and the soldier, for the 
characters were instantly interchangeable. Then too by 
ancient custom, while men prisoners might or might 
not be claimed by the general, women and children 
became the personal slaves of their individual captors. 

1 Lane- Poole, Stratford Canning, i, 395. His information came 
from a suspect Russian source. Bulwer, Palmerston, 1, 290. 



When the Turks were warring in Hungary, Constanti- 
nople abounded in Hungarian slaves; when Ibrahim 
was fighting in the Morea, Greeks filled the slave 
market of Cairo. This naturally shocked a generation 
which had so recently become alive to the horrors of 
the slave trade. But individual blame can scarcely be 
attached to Muhammad ’Ali or his son. “It must be 
remembered”, our consul-general wrote with complete 
but unwelcome truth, “that this is not a particular 
feature of the present contest. It is the same course that 
has been practised by the Turks in every war they have 

carried on Nor let it be supposed that it has been in 

the power as yet of the pasha to effect in this respect a 
change. It has only been by strict conformity with 
certain of these deeply rooted prejudices of his subjects 
that he has been able to accomplish so much.” Nor was 
the number anything like the myriads of gossip. About 
3000 had been brought to Cairo, mainly by speculators 
who had bought them from the troops, and of these 
about half had been liberated by various agencies. 
Some had been ransomed by European residents in 
Egypt; others had been released by their purchasers on 
the petition of their Greek servants ; and Muhammad 
’Ali himself had aided in this, sometimes by the issue of 
orders, sometimes by advances of money. 1 Such 
methods of war were barbarous ; it was high time they 
were suppressed ; but they were not personal or deliber- 
ate, and not nearly so extensive as was imagined. How- 
ever the truth ofa story has small relation to its influences 
upon the feelings, and the belief that the whole Greek 
race was being sold into bondage played a considerable 
part in leading the western powers to intervention. 

A better established fact led also in the same direc- 
tion. The Greeks of the islands had in general succeeded 
in repelling the attacks of Khusrau, the Capitan Pasha. 

1 Salt, August 12, 1826 (F.O. 78-147). 



But lack of money had created the greatest difficulty in 
keeping the Greek fleet together. Greek sailors were 
indisposed to serve even their country for nothing. 
Since they could not be paid, they must be suffered to 
plunder. They took to wholesale piracy under a thin 
disguise of blockade, seizure and condemnation. 
French ships bound for Candia with coin with which to 
pay for oil, were seized and their crews tortured to 
make them say where the money was. Hydra and 
Spezzia were the chief piratical centres. At Hydra the 
French admiral de Rigny sent an officer ashore to 
demand restitution of goods seized on a French vessel ; 
the populace at once gathered together, threatening to 
slay whoever dared betray the pirate. At Napoli was a 
so-called prize court. It was attended by the captors, 
pistol in hand, threatening to burn down the houses of 
judges who would not condemn their prizes. The 
commander of an Austrian squadron had to seize Greek 
vessels at Hydra and Spezzia to make good the losses of 
Austrian subjects. An English commander, similarly 
unable to obtain justice, entered and seized in the port 
of Hydra the pirates that he found there. 1 The Greek 
navy, which at first had displayed plenty of skill and 
gallantry, had thus dissolved into a disorganised body 
of corsairs, much more concerned with plundering 
European vessels than with destroying the Turks. 2 It 
was clear that if rebellion in the Turkish empire in- 
volved such interference with the freedom of the seas, 
and the Turks themselves could not keep Greek pirates 
in order, the powers whose trade was thus affected 
would be likely to intervene, in order to set a limit to 
the conflict. 

1 Douin, Navarin , pp. 3 sqq. 

2 They only made one attempt on Alexandria in 1827 and were 
easily repulsed. Driault, UExpidition de Crete et de la Moree, pp. 




The actual impulse however came neither from the 
outcries of the humanitarians nor from the piratical 
outrages of the Greeks, but from the political ambitions 
of Russia. The Emperor Alexander had always viewed 
his natural championship of the Orthodox Church as a 
convenient pretext for intervening in Turkish affairs. 
But in 1823-4 he was unwilling to break away from the 
other powers by any separate action, and accordingly 
planned, under cover of a European congress, a settle- 
ment which would have made Russia predominant in 
Greece. The projected congress was successfully evaded 
by Canning. But when at the end of 1825 Alexander 
died and was succeeded by Nicholas, other measures 
were needed to prevent the outbreak of war between 
Russia and Turkey over the Greek question. In these 
circumstances a joint Anglo-Russian mediation was 
proposed and accepted, and France was also induced to 
join. The result was the agreement of July 6, 1827, by 
which the three powers were to make combined efforts 
to induce the combatants to accept an armistice, and, in 
the event of their refusal, to “use all the means which 
circumstances may suggest to their prudence” in order 
to prevent further collision between them. The actual 
method contemplated was the blockade of the Morea, 
in order to starve Ibrahim out, by the naval forces of 
the three powers. 

Repeated representations had already been made to 
the Porte by the ambassadors, but had elicited no other 
answer than that the Greek rebellion was a purely 
internal matter of no legitimate interest whatever to the 
European powers. On August 16 the three dragomans 
carried to the Reis Effendi — the Foreign Minister — a 
note which he refused to receive. On the 29th they 
repeated their visit, and were assured that the Sultan 
would never accept any proposals regarding the Greeks 
and that he would persist in his resolve until the day of 





judgment. On the 31st they were sent again with a 
further declaration, which the minister, after a childish 
pretence of not understanding, again refused to accept. 1 
The only course then remaining open to the allied 
powers was the use of force. 

The cause of this demented decision was certainly a 
belief that Europe was too divided to intervene effec- 
tually, and that Russia would resent any action by the 
French and English squadrons. This belief seems to 
have been based in part on the difficult history of 
European coalitions; in part on the conduct of the 
Russian ambassador; 2 and in part on the deliberate 
suggestions of Austria. Metternich regarded the Greek 
rebels in the same light as Italian revolutionaries, and 
at the same time was convinced that the other powers 
would benefit more than Austria by intervention. The 
internuncio’s first dragoman held long and mysterious 
conferences with officials about the Sultan’s person; 3 
and we now know enough about the conduct of 
Austrian agents elsewhere to be certain that the inter- 
nuncio’s object was to urge the Sultan to make an end 
of the rebels as soon as possible. This of course jumped 
with Sultan Mahmud’s own instincts. The first hint of 
collective intervention had thrown him into an un- 
governable passion in which with tears of rage he had 
sworn to burn and ruin every province and city he had 
in Europe rather than submit to such intolerable hu- 
miliation. 4 And he had ordered his servants to declare 
that interference would only assure the extermination 
of the Greeks. “We will slay every Greek in all our 
territories, and when blood has begun to flow, so much 
the worse if the Armenians, our other rayas, and the 
Franks themselves mingle theirs with the blood of the 

1 Douin, Navarin , pp. 1 1 1 sqq. 

2 Idem , p. 1 1 7. 3 Idem, p. 12 1. 


guilty.” 1 Mahmud had yet to learn that he was not 
Sulaiman the Magnificent. 

Ideas of this kind certainly had no place in the plans 
of Muhammad ’Ali. His one purpose all along had 
been to strengthen his position either inside or outside 
the Turkish empire, with a decided preference for the 
second alternative if circumstances should permit. He 
had been much disturbed by the news that Lord 
Cochrane, that volatile admiral, had joined the Greek 
service ; 2 and he had listened in a very different manner 
from the Reis Effendi’s to the English remonstrances. 
He fancied indeed that he had found the lever needed 
to move the European world. 

Before ever the Greek revolt had begun he had made 
tentative overtures to England. In 1820 Salt was 
desirous of visiting London, partly for reasons of 
health, but partly also for reasons of state, “ our great 
man here having pressed me to make communications 
which I cannot commit to paper”. 3 Nothing came of 
this advance. But in 1826 the obvious reflection 
occurred to Stratford Canning at Constantinople that 
the easiest way of shaking Ottoman obstinacy would be 
to enlist the aid of the Pasha of Cairo. Would not, he 
wrote to Salt, a share in the tribute which at that time 
it was proposed that the Greeks should pay the Porte, 
and a pashaliq for Ibrahim in Syria be far more profit- 
able to Muhammad ’Ali than exterminating the Greeks 
at great cost to himself? 4 Salt at first thought that so 
flagrant a violation of Muslim sentiment as supporting 
the Greek cause could not possibly be looked for. 6 But 
a fortnight later began a series of conversations in 

1 Douin, Navarin , p. 122. 

2 Salt, November 1825 (F.O. 78-135), and August 4, 1826 

(F.O, 78-147). 3 Salt, August 20, 1820 (F.O. 78-96). 

4 S. Canning to Salt, June 10, 1826 (F.O. 78-147). Cf. Lane- 
Poole, op . cit. 1, 409. 

5 Salt to S. Canning, August 31, 1826 (F.O. 78-147). 



which the pasha gradually developed his views. He 
at once brushed aside all possibility of assisting the 
English cause at Constantinople. The Divan was too 
wavering, the Sultan too bigoted. But there were other 
ways of favouring our policy, and he would like to know 
what the British government would be prepared to 
offer him. A week later he reminded Salt that he had 
never put on his seal any other inscription than his 
name alone. “I have little left, as you may see, of the 
Pasha except my chiaoushes 1 with silver sticks and my 
divan.” He then dwelt upon the fact that geographi- 
cally and commercially “England and Egypt might be 
of use to each other — there is nothing that I desire 
more”. And when Salt broached the evacuation of the 
Morea, “That is not so easy a matter”, he answered; 
“it wants the assistance of some able helmsman to 
bring it about. But if there are those who desire it, 
they will be able to accomplish it.” In the last conversa- 
tion of the series, on September 26, he became more 
explicit. “ I now have my foot in two stirrups ... ”, 
he said. “Everything shall remain in balance as it now 
is until the spring. If in that time I find your govern- 
ment has any propositions to make that may please me, 
I shall be ready to embrace their offers, and will find 
means altogether to withdraw my troops from Greece. 
If not, I will collect all my force and through the in- 
fluence I have at the Porte get the command of the 
whole Ottoman fleet — for by that time the Capudan 
Pasha will have been disgraced; and then I will put 
myself at the head of it and so finish the business.” 
Salt asked what services he expected from England in 
return. Muhammad ’Ali mentioned help in enlarging 
his marine, and liberty to expand in Arabia ; but (Salt 
adds) “ I am persuaded that he has at heart the gaining 
from our government some general assurance of sanc- 
1 Corresponding with the Chobdars of Indian courts. 


tion to his independence, should any circumstances 
drive him into a rupture with the Porte, but he care- 
fully abstained from touching on this point”. 1 

Shortly after these conversations an Austrian diplo- 
mat reached Alexandria on a mission from Metternich. 
This was Prokesch-Osten, who was to make at a later 
time another peculiar visit to the pasha. On this 
occasion he was sent to spur on Muhammad ’Ali’s 
flagging activity and especially to press him to under- 
take a winter campaign against the Greeks, in order to 
secure their overthrow before Russia and the western 
powers should have time to intervene. He urged the 
danger of Greek independence to Egyptian commerce. 
He expatiated at length upon English desires to keep 
Egypt weak and averred that, however beneficial 
English advice might appear, its real object would be to 
paralyse, not to assist, the viceroy. But this argument 
failed to convince Muhammad ’Ali that any alliance 
could be nearly so beneficial to him as English friend- 
ship, or that any advantage could compensate for the 
hostility of English naval power. “ If England does not 
wish me to, what can I do?” he bluntly asked the 
Austrian. 2 

But as the weeks passed, and no response was made 
to his overtures, his mind naturally recurred to his 
alternative plan of securing from the Porte the whole 
control of the Greek war, and the more so because 
success at Constantinople would in no wise prevent him 
from coming to terms with the English, and would at 
the same time involve his personal enemy, Khusrau, in 
humiliation. He had already sent complaint on com- 
plaint of Khusrau’s misconduct in command of the 
Turkish fleet. 3 Now, on January 7, 1827, he wrote two 

1 Salt, October 1, 1826 (F.O. 78-147). 

2 Sabry, U Empire igyptien, pp. 1 20 sqq. 

3 E.g. to the Grand Vazir, Ramzan 5, 1241 (Abdine Archives). 



letters, one to the Grand Vazir, one to his personal 
agent 1 at Constantinople. In the first he said he had 
spared neither wealth nor men in the Sultan’s service ; 
but his resources were now exhausted, he himself was 
smitten by advancing age, and he therefore begged to 
be excused from further demands in order that he might 
spend his remaining years in peace, praying for his 
master’s health and prosperity. The significance of this 
humble petition was made clear by the second letter. 
It said, “Khusrau Pasha’s participation in affairs 
having been such as to produce neglect and inaction, if 
he is maintained in his office I shall cease from all co- 
operation with him and shall demand to be released 
from this service ”. 2 It so happened that shortly after 
these letters reached Constantinople, Stratford Canning 
made very unwelcome proposals to the Porte about the 
Greeks. In such circumstances procrastination was 
impossible even in the Turkish Divan. Almost at once 
an aga of distinction was ordered to proceed to Egypt 
on a secret mission. He sought a passage by a British 
man-of-war, being mortally afraid of being captured by 
the Greeks. Stratford Canning refused, being per- 
suaded that the mission would be unwelcome to the 
viceroy . 3 He need not have been anxious. The aga 
carried the news that Khusrau was no longer Capitan 
Pasha, and farmans investing Muhammad ’Ali with the 
sole conduct of the war. 

But with his customary prudence he was in no haste 
to withdraw a foot from either stirrup. He began 
leisurely preparations for the renewal of the campaign, 
but even at mid-June following his ships were still 

1 In the Turkish as in the Moghul empire, every great official at 
a distance from headquarters maintained an agent at court to 
watch over his interests. 

2 Douin, Navarin , p. 19. 

3 Stratford Canning, February 8, 1826 (F.O. 78-152). 



lying at Alexandria, his reinforcements for Ibrahim 
still incomplete. But he began to press our consul- 
general for an answer to his proposals, for his fleet could 
not be delayed for ever. Besides the Divan at Constanti- 
nople had observed that the change of command had 
made no difference to the slow progress of operations, 
and the wily Khusrau had already regained Mahmud’s 
favour and received the pelisse and sabre of invest- 
ment as sar-aska commander-in-chief of the Sultan’s 
forces. On June 1 1 Muhammad ’Ali assured Salt of his 
desire to comply with the wishes of the British and 
French governments. He suggested that, if the powers 
really intended to intervene, the British and French 
fleets had better appear before Alexandria “ to make a 
demonstration of compelling His Highness to desist 
from the war, and in that case, if they would show it to 
be to his advantage, he would immediately withdraw 
his troops and son from the Morea, His Highness 
assuring me that he wanted only a fair pretext for 
taking such a step”. 1 For nearly eight weeks longer he 
delayed despite the urgency of the Porte, despite the 
incitements of the Austrian consul. 2 At last, on 
August 6, the fleet put to sea. Two days later arrived an 
English envoy on a special mission. 3 

The new-comer was Major Cradock, whom Canning 
had sent specially to communicate the decision of the 
allies at London and to induce the pasha to abandon 
the Morea without more ado. He was to point out that 
the great powers were now agreed, that no intervention 
on behalf of the Turks could be expected, and that 
ample forces were being despatched to the Levant. If 
the Turks chose to resist, “the consequences of the 
Pasha’s identifying himself with the Porte in so unequal 

1 Salt, June 11, 1827 (F.O. 78-160). 

2 Douin, JVavarin, p. 150. 

3 Salt to Stratford Canning, August 12, 1827 (F.O. 78-160). 


a struggle might be fatal to those plans of maritime and 
commercial improvement which have hitherto been 
pursued by him with so much success”. These con- 
siderations, Canning thought, would be decisive with a 
man “wary, astute, neither a fanatical Mussulman nor 
a devoted servant of the Porte ”. 1 But though Cradock 
was expressly ordered to avoid the use of threats, his 
was not the most gracious of missions. As Salt observed, 
“we have to ask from him a neutrality which may com- 
promise him altogether with the Porte, and have 
nothing specifick to offer in return ”. 2 

A week 3 was spent at Cairo in discussions. The pasha 
was as amenable as could reasonably have been 
expected. Salt pressed him to seize this opportunity of 
explaining his wishes to the British government in 
precise terms, since “if such an opportunity of in- 
gratiating himself with the European powers were once 
permitted to slip by”, he could never expect another 
equally good. Muhammad ’Ali then advised that the 
allied admirals should formally require Ibrahim not to 
proceed to the attack of Hydra, which was the next 
projected operation, hinting that he himself would send 
corresponding orders. “Let England stand by me ”, he 
continued, “and I shall be repaid. I have long wished 
ardently ... to form a lasting league of commerce and 
amity with her, and she must, I should hope, now feel 
that she is bound to aid me.” In reply Salt gave it as 
his private opinion “that when the occasion came, 
should he carry this business successfully through, 
England would not desert him”. The pasha then 
burst out with his hopes of the future. His face lit up, 
his eyes flashed. “ Syria and Damascus and Arabia are 
in fact at my disposition. ... If your government 

1 Cradock’s instructions, July 14, 1827 (F.O. 78-182). 

2 Salt to Stratford Canning, August 12, 1827 (F.O. 78-160). 

8 August 15-21. 


support me, as I hope, if it will acknowledge me, when 
occasion comes, as an independent prince, I shall be 
satisfied.” 1 Before his departure Gradock himself, in 
conversation with Boghoz Bey, the pasha’s most con- 
fidential servant, gave it as his private opinion that if 
Egypt established and could maintain her independ- 
ence, it would be recognised by England. 2 

The discussions thus ended with no formal under- 
takings on either side. The viceroy hinted that his 
troops should be kept inactive in the Morea; the 
English agents said they thought that in that case he 
might rely on the benevolence of the British govern- 
ment. The pity was that Cradock had not been able to 
reach Alexandria in time to induce the pasha to defer 
the despatch of his fleet. His position had become most 
difficult. Pressed by the Sultan to exterminate the 
Greeks at once, pressed by France and England to 
withdraw at once from the Morea, it seemed that he 
could not but mortally offend one side or the other. 
He was himself convinced of the futility of further 
opposition to the wishes of the allies. But he was linked 
to a court too ignorant and proud to admit that the days 
were gone when the Sultan’s anger could safely confine 
western ambassadors in the Castle of the Seven Towers 
and when the Turks could meet the united force of 
Christendom on equal terms. On October 5 he made a 
strong endeavour to open the eyes of the Divan to the 
real position. He ordered his agent to observe that the 
demands of the powers might be mere bluff, but might 
also be enforced; that the wise prepare against mis- 
fortune instead of promising themselves prosperity; and 
that if the allied squadrons resorted to force, in his 
humble judgment the Ottoman fleet would be destroyed 

1 Salt, Memorandum, August 19, 1827 (F.O. 78-156). 

2 Cradock to Stratford Canning, August 21, 1827 (F.O. 78- 
182). Cf. Temperley, Foreign Policy of George Canning, pp. 148 sqq. 



and 30,000 or 40,000 souls perish. “It is wrong”, he 
continued, “in affairs of war to rely solely upon God. 
It is needful also to do all that can be done by man. 
Assuredly victory comes of God, and He alone controls 
all power and might. But in the Koran He has said : 
Strive and I will help you to win the victory.” 1 Faith, 
in short, was no remedy for damp powder or ill-found 

His prevision was unhappily justified. The allied 
admirals, Codrington and Rigny — the Russian squadron 
did not enter on the scene before October 1 3 — at once 
began to put pressure on the combatants. The Greeks 
of course immediately agreed to an armistice. But 
since the Sultan refused, they considered themselves 
released from the engagement, planned an expedition 
to Albania, and destroyed a Turkish flotilla at Galaxidi. 
Meanwhile the admirals themselves had had a personal 
interview with Ibrahim, who then agreed to suspend 
operations for a month until he could receive instruc- 
tions from the Porte and his father. But when he learnt 
of the continued Greek operations, he took action to 
revictual Patras and to clear the country in his occupa- 
tion of potential enemies. The admirals tried to hold 
the balance even. If on the one hand Codrington com- 
pelled the Turkish fleet to retire to Navarino without 
relieving Patras, so also on the other he forbade the 
Greeks to undertake their projected expedition into 
Albania. But he and his fellows were desirous of 
preventing further devastations in the Morea. They had 
nothing but naval force at their disposal. They there- 
fore sought to achieve their end by a combined demon- 
stration against the Turkish and Egyptian fleets. 2 On 

1 Douin, Navarin, pp. 243-5. 

2 Idem, chapters ix-xx. An account differing in some particulars 
from this will be found in Temperley, Foreign Policy of Camping, 
pp. 406-9. 



October 20 they sailed into Navarino Bay. The Turks 
inevitably distrusted their intentions, and when armed 
men fear attack the guns fire themselves. Musketry was 
opened on a British boat’s crew and it was answered 
with ship’s guns. From half-past two till nightfall a 
fierce action followed, in which the whole Muslim fleet 
was destroyed. 

This event was hailed with enthusiasm by all good 
Grecophiles. But it surprised the allied governments. 
They had in fact tried to do with naval force more than 
naval force alone can possibly accomplish. Its effect on 
land operations is slow, constrictive, gradual ; and what 
the allies wanted was an immediate cessation of arms. 
They were giving the admirals an impossible task. 
Again their instructions were faulty and incomplete, as 
a result of the illogical position which they had 
assumed. While they posed as intervening between the 
Sultan and his rebel subjects, they were in fact inter- 
vening to save the Greeks; so that while the action 
certainly went beyond the expectations of the western 
powers, it also with equal certainty helped strongly 
forward the accomplishment of their purpose. But to 
Muhammad ’Ali it must have been a double blow. He 
had been ready to negotiate with the allies, and, had 
Cradock only arrived two days earlier, his fleet, as 
Canning expected, might never have sailed, and 
Navarino would never have been fought. Ibrahim and 
the Divan at Constantinople were at first in favour of 
retirement northwards out of the Morea and beyond 
the immediate influence of the allied vessels. But 
Muhammad ’Ali was in no mind to continue the fruit- 
less conflict. The day after the news was received he 
informed the English consul that even if war broke out 
between Turkey and Great Britain, British subjects in 
Egypt should be unmolested. “I know well”, he said, 
“how to appreciate and to maintain the reputation I 



have acquired for justice and liberality.” 1 The same 
day he wrote to his son declaring that the Divan’s 
stupidity had been the cause of the misfortune, and 
that he was to remain in camp without making any 
attempt upon the Greeks. 2 When he heard of the 
proposal to move Ibrahim’s army to the northward, he 
protested strongly and effectually. 3 Ibrahim therefore 
remained in the Morea until the debarkation of a 
French force made his position so desperate that even 
the Porte could not but recognise the cogency of his 
capitulation. On August 6, 1828, Codrington visited 
Alexandria and signed the convention with Muhammad 
’Ali by which the Morea was finally evacuated. 4 * Since 
the Sultan still remained obdurate, the Russians pro- 
ceeded to employ military force, and in the following 
year, when they forced him to sign the Treaty of 
Adrianople, he at last acquiesced in the views which the 
viceroy of Egypt had submitted to his ministers two 
years before. 

This excursion into European politics had strained 
Muhammad ’Ali’s resources to the utmost. The money 
he had lavished on his ships, the victuals and munitions 
which he had poured into the Morea, the men whom he 
had levied, trained, and sent, had been wasted. Ibrahim’s 
army returned starved, crippled and miserable. Many 
were so worn by privations that they could not march. 6 

So far the Divan at Constantinople had succeeded in 
putting the Greek rebellion to good use. The strong 
pasha was less strong than he had been before he under- 
took the office of champion of Islam. But he had not 
been satisfactorily embroiled with the western powers, 

1 Barker, Syria and Egypt, 11, 58. 

2 To Ibrahim, Rabi-al-awal 13, 1243 (Abdine Archives). 

3 To Najib Effendi, Jumadi-al-awal, 1243 {ibid.). 

4 Convention dated August 6, 1828 (F.O. 78-170). 

® Barker to Sir P. Malcolm, September 24, 1828 {ibid.). 



though they had burnt his fleet and starved his army. 
He had unluckily had sufficient insight to lay these mis- 
fortunes at the door of the “pig-headed Sultan” and 
“ass-like Vazirs ” who had rejected his advice. He had 
withdrawn from the struggle and looked on unharmed 
while the conquering Russians had been daily expected 
at the very gates of Constantinople. He was justly 
filled with contempt for the impotence and malice of 
the Sublime Porte, and more resolved than ever to 
establish his complete freedom from its blighting in- 
fluence. He was more convinced than ever of the 
efficacy of sea power, and especially of British sea 
power. The possession of the Morea indeed had not 
proved to be the pledge which would purchase for him 
a British alliance because in a single afternoon sea 
power had torn it from his grasp. But the necessary 
pledge he still might find. Might it not be the mastery 
of both the overland routes to India? Might it not be 
the threat of an alliance with Britain’s great Mediter- 
ranean rival? 

Chapter IV 


Along the north African coast from Mogador to Ben- 
ghazi lay the Barbary states, piratical principalities 
that once had formed part of the caliphate, but with the 
break-up of the Islamic world had attained a virtual 
independence. The rise of the Ottoman empire had 
left them almost free, though attached by ties of envious 
veneration and respect for the great barbaric power 
which had seated itself at Constantinople. All through 
the sixteenth and seventeenth centuries they had 
carried on a ceaseless war against all Christian mariners. 
They had provided the excuse for ship-money, and 
every English boy was familiar with the capture and 
enslavement of Robinson Crusoe by the Moors of 
Sallee. The rise of the western navies in the eighteenth 
century had somewhat abated their predatory activi- 
ties ; but their inclinations remained quite unchanged, 
and though they became somewhat shy of meddling 
with French or English vessels, they still preyed freely 
enough on Spanish, Genoese or Neapolitan ships that 
fell within their reach. Between 1805 and 1815 they 
took some ninety ships, and though discouraged by 
Lord Exmouth’s bombardment of Algiers in 1816, they 
nevertheless captured twenty-six more in the ten years 
that followed. In 1824 Algiers, the chief offender, was 
again visited by an English fleet. The time was evi- 
dently drawing near when these upholders of medieval 
tradition would be brought to a sharp reckoning. 

Like all good Muslims, the Barbary Moors had been 


deeply moved by Christian interference in the affairs of 
Greece and had lent the aid of their shipping to the 
Sultan, resenting the lost freedom of the seas and by no 
means foreseeing the catastrophe of Navarino. In their 
sore, irritable frame of mind they were disposed to 
defy the Occident and all its fleets. In April, 1827, 
Husain, the Dey of Algiers, in the course of a stormy 
interview with the French consul, Deval, struck the 
consul with his fly-whisk. Reparation was demanded 
and refused. The consul was withdrawn, and Algiers 
was blockaded by a French squadron. As the Dey 
remained impenitent, and the general situation (especi- 
ally the outbreak of the Russo-Turkish war in 1828) did 
not permit of more vigorous action, attempts were made 
by the Sardinian consul and then by a French naval 
officer to persuade him to accept easier terms than at 
first had been demanded. This convinced the Dey that 
the French were weakening and made him more 
obstinate than ever. Then in the middle of 1829 the 
Provence was sent under a flag of truce with new pro- 
posals, coupled with threats of an expedition if they 
should not be accepted. But Husain still refused. 
When threatened with attack, “I have powder and 
guns,” he said, “ and since we cannot agree, you may 
depart”. On August 3, therefore, the Provence still 
flying her flag of truce set sail. The wind carried her 
past the town batteries. The Moors, perhaps fancying 
that this was a deliberate insult, opened fire on her and 
maintained it as long as she remained within range. 
With an expenditure of eighty rounds they actually 
succeeded thrice in hitting her. 

When this news reached Paris, public opinion, already 
indignant at the delay in bringing the Dey to terms, ran 
dangerously high. But the moment was strangely em- 
barrassing. The Russians had just occupied Adrianople, 
the collapse and partition of the Ottoman empire 


seemed at hand. Could a prudent minister at such a 
moment commit the naval and military forces of 
France to a campaign in northern Africa? Besides 
Polignac, who had become minister for foreign affairs 
in August, had just formulated a scheme which, could 
it but be effected, would restore all the popularity of the 
tottering throne of Charles X and undo the work of the 
allies on the overthrow of Napoleon . 1 He assumed that 
Russia and Austria would divide between them most of 
the Turkish provinces in Europe, and so open the way 
for France to demand an equivalent for the disturbance 
of the balance of power. His project was that France 
should receive the Belgian provinces up to the line of 
the Meuse and Rhine. The assent of Prussia might be 
secured by letting her annex Saxony and the northern 
Dutch provinces. The King of Holland should be com- 
pensated for this partition of his kingdom by being sent 
to reign at Constantinople over .such of the Turkish 
provinces in Europe as were not absorbed by Russia 
and Austria, while Great Britain should be offered as 
the price of acquiescence the Dutch colonies which 
would by this arrangement have become res nullius. 
This plan was to be carried into action by a secret 
treaty between France and Russia; that signed, Prussia 
would be invited to join; Austria then would have no 
choice but to fall in with the scheme, and Great Britain 
could accept or reject Java and the Moluccas as she 
pleased. As soon as the treaty had been signed, the 
agreeing powers would assemble their troops, so as to 
confront Europe with such a force as none of the re- 
maining states could think of resisting. Polignac 
thought of putting 200,000 men into the field. But 
such ideas absolutely ruled out all possibility of any 

1 He was no doubt deeply influenced, not only by the un- 
popularity of the king, but also by the active discontent of the 
Southern Netherlands. 


immediate expedition to chastise the peccant Dey of 

In these circumstances the French minister resolved 
to adopt a plan which had been suggested by Drovetti, 
who had long served as consul and consul-general in 
Egypt, and who in 1829 had just returned to France. 
This new plan was to punish the Dey by the hand of 
Muhammad ’Ali, who would organise a great expedi- 
tion to conquer and annex the three Barbary regencies 
of Tripoli, Tunis and Algiers. A direct French expedi- 
tion, Drovetti argued, would provoke English jealousy 
and opposition, whereas the extension of the pasha’s 
authority along the African coast would not offer the 
same opening to diplomatic protests, and besides that 
(Polignac may well have thought) the proposed re- 
vision of the European map would keep the English 
cabinet too busy for it to spare a thought to what 
became of Tunis and Algiers, while every other Euro- 
pean state would certainly welcome the establishment 
in those regions of the good order and discipline which 
reigned in Cairo and Alexandria. 1 

This plan seems to have originated with Drovetti 
himself. He had urged on Muhammad ’Ali the advan- 
tages of co-operating with the French in Algiers instead 
of alarming all Europe with adventures in Syria; 2 and 
he seems to have assumed that the advantages of his 
proposal would be as evident to the English as they 
were to himself. By 1829 he was so full of the plan that 
it had become his favourite subject of conversation, 
even with the English consul-general, Barker. The 
latter, in the seclusion of his own office, judged the 
project to be wholly chimerical; but while he was 
listening to the enthusiastic speeches of Drovetti all 
difficulties vanished, and French support in ships and 

1 Douin, Mahomet Ali et V Expedition d’ Alger, pp. 1 sqq. 

2 Idem, p. 6. 




men seemed merely to be making success doubly 
certain . 1 Muhammad ’Ali’s own attitude was perhaps 
not entirely that which he led Drovetti to suppose. He 
was not really interested in the Barbary states, and 
probably was well aware that an extension of his 
authority in that direction would be a source of weak- 
ness rather than of strength. He fully realised the 
strategic importance of the area comprising the pasha- 
liqs of Syria and Baghdad, and well knew that if he was 
to rise to the position of weight and power which he 
coveted, Syria and Baghdad would be far more valuable 
than the north African littoral. But at the same time he 
never liked to let chances slip away unused. If the 
French proposals ever came to anything, they might 
afford him two advantages — an opportunity of re- 
building his vanished fleet, and the possibility of an 
alliance either with the French themselves, or, if that 
prospect was sufficiently alarming to the English, then 
with Great Britain. He was in fact willing to undertake 
the conquest of Algiers if it was made worth his 
while or to lay aside the scheme if that would pay him 

Drovetti, it seems, was too deeply enamoured of his 
plan to read aright the pasha’s intentions; while 
Polignac was eager to adopt any plan which would at 
once placate French opinion by inflicting due punish- 
ment on Algiers and conserve French forces for the 
much more momentous European project simmering 
in his mind. He therefore laid the matter before the 
king, secured the royal assent, and then, apparently 
without consulting his colleagues, issued instructions to 
Guilleminot, the ambassador at Constantinople, and 
Mimaut, the new consul-general at Alexandria. The 
first was ordered to obtain farmans authorising Muham- 
mad ’Ali to subdue the three Barbary regencies; this 
1 Barker, August 18, 1829 (F.O. 78-184). 


demand was to be supported by the arguments, first 
that if the French themselves despatched a punitive 
expedition they would probably be obliged to keep the 
territory which would thus be lost for ever to the Porte, 
and secondly that Muhammad ’Ali would pay tribute, 
which the Deys never did . 1 The second was instructed to 
inform the pasha that France approved of his views and 
favoured his designs against the Barbary states, that if 
he wished it the French fleet should co-operate with his 
forces, and that he should receive an advance of ten 
million francs if the expedition set out without delay . 2 

But the discussions at Constantinople and Alexandria 
did not go so smoothly as Polignac had over-hastily 
assumed. Muhammad ’Ali strongly deprecated open- 
ing the matter at Constantinople at all, urging that the 
Sublime Porte would never willingly permit any exten- 
sion of his power, and might seek the assistance of an 
English fleet to counteract his operations, whereas if it 
were not consulted beforehand it would bow easily 
enough to an accomplished fact . 3 These objections 
proved well founded. In vain did Guilleminot, the 
ambassador, make the most of all the advantages of the 
French plan. As the viceroy had foreseen, the Porte, 
though not of course avowing it, had the most rooted 
objection to anything likely to increase its ambitious 
vassal’s power or prestige. It put forward a counter- 
proposal. All that was needed to end the dispute 
between the French and the Dey, declared the Reis 
Effendi, was the interposition of the Sultan’s exalted 
authority, and for this purpose he offered to despatch 
an agent of his own, Tahir Pasha, a known and bitter 
enemy of the French, who would without question 

1 To Guilleminot, October 10, 1829 (Douin, Mahomet Ali et 
V Expedition d' Alger, pp. 9 sqq.). 

2 To Mimaut, October 19, 1829 (Douin, op. cit. pp. 14 sqq.). 

3 Mimaut, November 27, 1829 (Douin, op. cit. pp. 23 sqq.). 



bring the Dey to reason without any need of force. 1 
And while this obstructive suggestion was under dis- 
cussion, the Turkish minister informed the English 
ambassador, Sir Robert Gordon, of what was going 
forward, shrewdly calculating that this was the best 
method of defeating plans which the Divan intensely 
disliked. 2 

At Alexandria Muhammad ’Ali declared he was 
willing to despatch 20,000 regular troops, and a like 
number of Bedouin, under the command of his son 
Ibrahim, but he required an advance of at least twice 
as much as the consul-general, Mimaut, had been em- 
powered to offer, and above all he demanded as an 
essential part of the bargain that the French should 
give him under cover of a pretended sale four 80-gun 
men-of-war, which he declared essential to the prompt 
success of the expedition and the avoidance of incon- 
venient foreign intervention. All the efforts of Mimaut 
and of Huder, who had been specially sent to aid 
Mimaut in the negotiations, failed to induce him to 
depart from his demand of four ships, which he said had 
formed an integral part of his discussions in the past 
with Drovetti. 3 

Huder returned therefore to Paris to lay these require- 
ments before Polignac, who was made acquainted with 
them on December 26. At this time, although the 
Peace of Adrianople had put the general plan for the 
modification of European boundaries out of the question, 
Polignac still hoped to secure Russian support for the 
annexation of the Belgian provinces. 4 He therefore 
decided to accept the proposals of Muhammad ’Ali, 

1 Guilleminot, December 9, 1829 (Douin, Mahomet Ali et 
VExpSdition d’ Alger, pp. 53 sqq.). 

2 Gordon, December 15, 1829 (F.O. 78-181). 

3 Mimaut, November 27, 1829 (Douin, op. cit. pp. 27 sqq.). 

4 Douin, op. cit. p. xxviii. 


and laid them before his colleagues. But here he met 
with strong opposition. More than one member of the 
cabinet felt it would be equivocal and perhaps dis- 
honourable to transfer vessels flying the French flag to 
another state. The Minister of Marine was obstinately 
opposed to any such weakening of the French navy, 
and threatened to resign should the proposal be ac- 
cepted. Bourmont, the Minister of War, saw prospects 
of personal glory in a possible expedition to Algiers and 
was most unwilling to allow himself to be replaced by 
Ibrahim Pasha. After repeated efforts Polignac could 
not induce the cabinet to go further than to consent to 
an advance of twenty-eight millions — twenty to be 
handed over to Muhammad ’Ali in accordance with his 
demand, and eight to be devoted to building expressly 
for him four men-of-war such as he desired. But a 
French squadron was to be held in readiness to support 
Ibrahim’s operations in case it was required. Huder 
therefore returned to Alexandria with these revised 
terms, the commander of the French squadron in the 
Levant was ordered to prevent the Turkish fleet from 
threatening Alexandria or attacking any Egyptian trans- 
ports proceeding against the Barbary states, and, 
since the time of overt action was now drawing near, 
Polignac at last opened the matter to the states of 
Europe. 1 

In spite of the secrecy which had been observed at 
Paris, the English cabinet was not ignorant of the pro- 
jects that had been formed. Barker had reported 
Drovetti’s conversations in 1 829. Gordon at Constanti- 
nople had sent home the interesting communication 
which he had received from the Reis Effendi. Metter- 
nich had intercepted French despatches from Guille- 
minot and made haste to favour our ambassador, Lord 
Cowley, with copies. Meanwhile the French ministry 
1 Douin, Mahomet Ali et V Expedition <T Alger, pp. xxx sqq. 


had replied to all the English enquiries with absolute 
denials. This conduct was little likely to induce Aber- 
deen and Wellington to accept the statements at last 
made to them or to acquiesce in the ostensible policy at 
last declared. On January 23 the Due de Laval (the 
French ambassador at London) called on Wellington 
and read to him an ostensible despatch from Polignac. 
The ambassador was received with marked coolness. 
It was observed that Muhammad ’Ali could not legiti- 
mately take up arms against the Barbary states except 
on behalf and by order of his master the Sultan, and 
a hope was expressed that the French ministry would 
abandon its plan of common action with the viceroy of 
Cairo. 1 Aberdeen immediately wrote to the English 
representatives at Cairo and Constantinople. To the 
latter he declared that whether the Sultan had or had 
not authorised the project, we could not view with in- 
difference such a change in the possession , of important 
African territories effected by French means, under 
French influence, and presumably in order to promote 
French interests. 2 To the former, after dwelling on 
British objections to seeing the pasha undertaking such a 
scheme under French patronage, Aberdeen added that 
he hoped Muhammad ’Ali would not doubt the friend- 
liness of the motives which impelled the British govern- 
ment “to counsel him on this occasion to weigh well 
the serious consequences of the enterprize in which he 
seems disposed to engage”. 3 

This opposition to the French project could have 
surprised no one. The establishment of French in- 
fluence in Algiers, either directly or through some third 
party acting on their behalf, would change the position 
in the Mediterranean and the question of protecting 

1 Douin, Mahomet Ali et l’ Expedition d’ Alger, pp. xli sqq. 

2 To Gordon, January 25, 1830 (F.O. 78-188). 

3 To Barker, January 29, 1830 (F.O. 78-192). 


British interests would arise. But there were much 
greater possibilities involved in the proposal. Beyond 
the status of Algiers loomed the whole Eastern Question. 
If Muhammad ’Ali conquered Algiers for the French, 
that could mean nothing else than that he would be 
taken directly under French protection. His position 
as regards his sovereign the Sultan would be affected. 
The status of Egypt would be tacitly modified. The 
tottering walls of the Turkish empire would receive 
another shock, and the Ottomans would become less 
than ever capable of opposing their Russian neigh- 
bours. To the French government, now as in the past, 
such events might be welcome as affording a possible 
opportunity for the expansion of French power; Poli- 
gnac had just been seeking to turn them to the advantage 
of the monarchy. But in English eyes they were preg- 
nant with danger. One of our major interests demanded 
that no European state should be suffered to establish 
itself on the routes to India. To France then the main- 
tenance of the Turkish empire was an open question to 
be decided in the light of interests mainly continental ; 
but to Great Britain it was the only alternative to de- 
velopments which no one could forecast and which 
therefore it was but wise to retard as long as possible. 
The position in 1830 thus anticipates and points on to 
the much sharper clash of French and English policy 
that was to come ten years later. 

But at the moment this firm declaration of English 
disapproval coincided with the complete failure of 
Polignac’s plans to recover the Rhine frontier. His 
secret and obscure discussions with St Petersburg — a 
special cipher was used and destroyed afterwards 1 — 
came to nothing. Prussia bluntly answered that nothing 
would induce her to let France advance to the left 
bank of the Rhine. The forces that till then had been 

1 Douin, Mahomet Ali et V Expedition d 9 Alger, p. xxviii. 


immobilised by the possible need of supporting the 
European project were thus set free just when it became 
clear that the whole weight of English influence would 
be thrown into the scale against Muhammad ’Ali’s 
occupation of the Barbary states. Polignac therefore 
resolved once more to change his plan, to limit Muham- 
mad ’Ali’s part to the conquest of Tripoli and Tunis, 
and to send a French expedition against Algiers. Thus 
the viceroy of Egypt would still be made an active ally 
of the French, ready to be recognised (as Polignac him- 
self wrote) at a convenient time pour le lieutenant du Roi 
de France 1 

But in this he was reckoning without his proposed 
ally. From the first Muhammad ’Ali had never in- 
tended to move unless he secured advantages, naval 
and political, which would make it clearly worth his 
while. Nor can he have been favourably impressed by 
the wavering conduct of French policy at this time. 
Resolute himself, he thought slightly of those who 
shifted their designs from day to day. The inconstancy 
of French plans probably led him to distrust the value 
of a French alliance, which, whatever advantages it 
might secure, would assuredly be accompanied by an 
unwavering English hostility. Before the despatch of 
Aberdeen had been communicated to him, he had 
already rejected the French proposals . 2 

A few days later he received in audience the English 
consul-general who had gone up especially from Alex- 
andria to Cairo to inform him of Aberdeen’s views. The 
viceroy objected that the English warning was super- 
fluous, and then proceeded once more, as he had done 
to Salt, to expound his views and declare his desire for a 
friendly understanding with Great Britain. “ Do you not 

1 Douin, Mahomet Ali et I’Expedition d’ Alger, p. Iviii. 

2 Mimaut, March 2, 1830, and Huder, of the same date 
(Douin, op. cit. pp. 191 sqq. and 197 sqq.). 


see that it is impossible to maintain the Porte?” he 
demanded. “You may prop here and prop there, but 
all will be to no purpose. What can you do with a 
government that has lost the confidence of the people 
both within the capital and in the provinces?. . .” It 
was therefore idle to look to the Turks for any effective 
resistance to future Russian aggression. Yet that was a 
matter in which English interests were most deeply in- 
volved. “The only way to strengthen the Sultan is to 
support me. By supporting me, he would soon have at 
his disposal a disciplined army of 125,000 men, ready 
to form a barrier against the Russians both at Con- 
stantinople and in Persia. In Persia, it is there after all 
that England must fight the Russians.” What is the use 
of peeping through your fingers and pretending to see 
nothing? “The Porte is gone, and England must pre- 
pare to raise a force in Asia to meet the Russians. And 
where can she find it but with me and with my son 
after me?” 

He then dwelt at length on the ease with which the 
Ottomans would join his standard “if the English 
would but come forward and support me”, and upon 
the extent of his resources, which he said with truth the 
English ministry underrated. And then he burst out, 
“With the English for my friends I can do anything : with- 
out their friendship I can do nothing. ... I foresaw long 
ago that I could undertake nothing grand without her 
permission. Wherever I turn, she is there to baffle me ”. 1 

Statesmen seldom utter more of their inner thoughts 
than they think well to make known to their audience, 
and Muhammad ’Ali was no child to blurt out more 
than he intended. But he was undoubtedly sincere in 
his statement of his attitude towards England. She 
enveloped him on every side. No other power could 
aid him so effectually. There was great truth too in his 
1 Barker, March 8, 1830 (F.O. 78-192). 


estimate of his own position and opportunity. He 
represented at the moment the one live progressive 
force in the world of Islam. He might with English aid 
have built up under the shadow of the Turkish cali- 
phate such a power as our own East India Company 
had built under the shadow of the Delhi empire. But 
once more what advantages could he offer to induce 
English statesmen to abandon their declared policy and 
co-operate to establish a new power in the Near East? 
Support against Russian designs? Perhaps. But were 
he once firmly entrenched on the borderlands of 
Persia, with his power stretching from Cairo to Bagh- 
dad, might not the Russians have something to offer 
which would induce him to forsake his English friends? 
And then our position would be dangerous indeed. 
Only some most pressing motive could warrant our 
supporting Egypt in a career of conquest. Probably a 
great European crisis alone could provide such a 
motive. In any case the trend of English policy was 
decidedly opposed, not, as some writers have thought, 
to Egyptian greatness, but to Egyptian intrusion into 
areas beyond the natural limits of the country. 

Muhammad ’Ali had long coveted the four pashaliqs 
of Syria. Their possession would cover his Egyptian 
territory from attacks by the Turks ; would give him the 
rule of Jerusalem, another of the Holy Cities of Islam, 
and so enhance his prestige in the Muslim world; and 
as he supposed, though the event proved otherwise, 
increase his resources in men and money. It would 
give him Damascus, one of the leading centres of 
Islamic culture. It would give him control of territory 
rich in timber and save his buying it at high prices in 
Trieste. And it would demonstrate to the world at 
large his favourite thesis of the vanished power of the 
Porte and his unique capacity for reorganising and 
rejuvenating Turkish power. 


The pashaliqs were in a miserable condition. They 
were so disturbed that couriers could not pass through 
them with any certainty of safety. 1 They had long been 
governed by pashas whose rapacity had been limited 
only by their power. No one ventured the least parade 
of wealth. Everyone was or at all events seemed miser- 
able, and the mixed population with its various sects and 
creeds was seamed with ineradicable feuds and quarrels. 

This land of misrule had been desired by the viceroy 
for many years. As early as 1812 he had spoken to the 
English consul of his intention to conquer Palestine as 
soon as circumstances should permit. 2 For the time 
being, however, he found too many obstacles in the 
way. The chief one was perhaps the need of reorganis- 
ing his army so as to render it a trustworthy instrument 
of his will. And then also the spiritual influence of the 
Sultan had to be considered, especially in those years of 
rising fanaticism which coincided with the Greek 
revolt. “Such are the religious prejudices of these 
people”, he said to Salt so late as the year ofNavarino, 
“that they all desert a pasha when once under inter- 
diction of the head of the Church.” To oppose the 
Sultan with effect, he added, a pasha “must be strong 
enough to command publick opinion and that is not an 
easy matter”; and this view he supported by the 
example of a rebel pasha in Kurdistan, whose troops 
fell from him “like sand from a pilgrim’s feet”. 3 

But in 1 830 the Egyptian army had been reorganised ; 
Ibrahim had been proved to be a skilful and resolute 
leader; while conscription promised as many men as 
should be wanted. At the same time the disasters 

1 Cartwright to the East India Company, November 9, 1822 
(I.O. Egypt and the Red Sea, 7). 

2 Missett, June 20, 1812 (F.O. 24-4). 

3 Salt, Memorandum, January 20, 1827, enc. in despatch of 
February 10, 1827 (F.O. 78-160). 


which had befallen the Turks at the hands of the in- 
fidels, afloat at Navarino, and ashore in the Russian 
campaign, showed even to thick-headed Turks that 
Sultan Mahmud was no certain guide to victory. The 
empire needed indeed only one good push to over- 
throw it altogether. 

And while the negative reasons which had earlier 
held Muhammad ’Ali back had vanished, an addi- 
tional and positive reason had appeared. The bait by 
which he had been lured into taking part in the Greek 
war had been the four pashaliqs of Syria. They had been 
promised; but with the revival of Khusrau’s influence 
at the Porte the promise had been cast aside. At the 
close of 1827 the viceroy was still demanding in vain the 
farmans of investiture. 1 He had therefore lost his fleet, 
and imperilled his army and his son for nothing. He 
made up his mind to occupy Syria before anyone else 

Pretexts were of course not lacking. He had been 
called upon by the Porte to provide assistance in sup- 
pressing the rebellion of Mustafa Pasha Iskudarli in 
Rumelia. This demand covered his military prepara- 
tions; and when he was informed that help was no 
longer needed, he proposed to use his assembled forces 
to attack Abdullah Pasha of Acre, who had been levying 
exactions on Egyptian merchants. 2 Another excuse 
was the welcome which Abdullah gave to the fellahin 
who fled to Acre to escape Muhammad ’All’s con- 
scription. In the course of 1831, 6000 are said to have 
escaped thus. Abdullah refused to return them. Mu- 
hammad ’Ali replied that he would come and take 
them. 3 He set his troops in motion in October, 1831. 

1 Salt, August 27, 1827 (F.O. 78-160). Muhammad ’Ali to the 
Shaikh Effendi, Jumadi-al-awal 23, 1243 (Abdine Archives). 

2 Grand Vazir to the Wali of Damascus, Rabi-al-awal 3, 1247 

{ibid.). 3 Sabry, op. cit. p. 191. 


A singular commentary on the incompetence of the 
Divan at this time is afforded by its reception of 
Muhammad ’Ali’s proposal to attack Abdullah. The 
Grand Vazir was aware that he was preparing to 
occupy the Arabic-speaking parts of the empire, to 
organise their affairs, and proclaim his independence, 
but no better expedient could be devised to meet the 
danger than to warn Abdullah to be circumspect and 
avoid provoking a quarrel, while a bland letter was 
addressed to Muhammad ’Ali himself, telling him that 
a few merchants’ complaints did not justify a declara- 
tion of war and that disputes between neighbouring 
pashas should be determined, not sword in hand, but 
by the mediation of the Sublime Porte. 1 But nothing 
was done, no preparations were made, no orders were 
issued, to guard against the impending danger. 

In accordance with the Egyptian plans Acre was 
besieged by sea and land. But the Egyptians en- 
countered a stout resistance. Abdullah may not have 
been very honest or very wise, but he was a brave and 
resolute man, while the siege in its early stage was con- 
ducted with great want of skill. On December 9 an 
attempt was made to overwhelm the place with a com- 
bined bombardment from the shipping and the land 
batteries. But the ships suffered considerable damage 
and the land batteries made small impression. After 
three months more of effort a determined attempt was 
made to storm the walls. It almost succeeded. A party 
of the stormers penetrated to the market place, but 
finding themselves unsupported had to withdraw. 
Ibrahim’s position began to appear critical. 2 Various 
bodies of troops were assembling to relieve the town, 
and the Porte, gathering courage from its protracted 
defence, ventured to strike out the names of Muham- 

1 Grand Vazir to the Wali, ut supra. 

2 Barker, Syria and Egypt, ix, 1 79-80. 


mad ’Ali and Ibrahim from the list of the pashas of the 
empire published annually at the Feast of Bairam, 
which in 1832 fell at this moment. A feeling of unrest 
spread through Cairo and AJexandria. Men began to 
murmur against the viceroy’s government. On March 
14, and again on the 21st and 23rd, men found exposed 
by the Bab-uz-Zuwaila at Cairo the newly beheaded 
bodies of three Turks, two of them soldiers and the 
third one of the ulema, with labels on their breasts de- 
claring, “This is the fate which awaits those who 
cannot govern their tongues”. 1 And on April 7 two 
new corpses were exposed, with the grim warning, 
“This is the punishment that awaits those who speak 
against the government in secret”. 2 

The malcontents, however, were reckoning not only 
without the spies of Muhammad ’Ali but also without 
the unskilled commanders of the Turkish forces. After 
the failure of his attack on March 9, Ibrahim deter- 
mined to leave some 5000 men to blockade Acre, while 
with the remainder he moved against his gathering 
enemies. After scattering 12,000 Turks who had con- 
centrated near Homs, Ibrahim then returned to renew 
his attack upon Acre. At dawn on May 27 he directed 
the storming party in person. A long struggle ensued. 
Ibrahim is related to have cut down with his own hand 
several officers who had fallen into the rear of the 
storming columns, and after great efforts he had by 
nightfall succeeded in capturing the place, which was 
then, in strict accordance with western rules of war, 
given up to plunder. 3 Abdullah proudly declared: “I 
had walls, men, and money, with which to defend 
Acre. When Ibrahim took it the walls had been 
destroyed ; of my 6000 men 5600 were dead, and of my 

1 Barker to Stratford Canning, March 29, 1832 (F.O. 78-213). 

2 Same to same, April 11, 1832 (ibid.). 

3 Barker, June 4 and 27, 1832 (F.O. 78-214). 


treasure nothing remained but a few jewels”. He laid 
with good reason the blame of his defeat upon the 
Sultan who had done nothing to relieve him. “The 
Porte”, he said bitterly, “has the honour of a dancing- 
girl”. 1 

Acre having fallen, Ibrahim once more advanced 
northwards. He entered Damascus without opposition 
on June 13. On July 8 he unexpectedly attacked the 
Turkish forces at Homs, and after a brief action com- 
pletely routed them, capturing their artillery and 
baggage. On the 15th he occupied Aleppo. On the 
29th he defeated another Turkish force in the Beylan 
Pass. With that for a while action was suspended. 

Clearly enough two courses were now open to the 
viceroy. He might proclaim his independence and 
press his advance against the demoralised Turks in 
order to force the Sultan to recognise the position ; or he 
might pause in the hopes of obtaining by western inter- 
vention such a settlement as he desired. There were 
grave dangers attached to either plan. A direct advance 
on Constantinople, which was advocated by Ibrahim, 
was disagreeably likely to provoke a hostile interven- 
tion in favour of the Sultan. When Ibrahim proposed to 
strike coin and have the Friday prayers read in his 
father’s name, Muhammad ’Ali refused. He had 
attained to power by moderation, he declared, and 
would have no new titles or honorifics. 2 While Ibrahim, 
carried away by his triumphs, fancied that everything 
they wanted could be secured by defeating the Turks, 
his father shrewdly remembered that other and more 
powerful states than Turkey had to be dealt with if his 
success was to be consolidated. Ibrahim’s advance to 
Constantinople would be the certain signal for action 
by the powers that had already intervened in Greece. 
A pause in the advance would certainly allow the 
1 Sabry, op. cit. pp. 197-8. 2 Idem, p. 205. 


Turks to recover from their panic, to reassemble their 
forces, and to guard the road to Constantinople. But 
the Turks had been beaten once and could be beaten 
again. Muhammad ’Ali, reckoning Turkish soldiers a 
much less formidable enemy than France and Britain, 
preferred therefore to pause and negotiate. 

Indeed at the moment the attitude of both powers 
was friendly. The monarchy of July desired to see 
Muhammad ’Ali’s power strengthened, so long as 
that involved no convulsion at Constantinople serious 
enough to raise prematurely the question of the parti- 
tion of the Turkish empire. Accordingly from the 
middle of 1832 all the French influence was directed 
towards inducing Muhammad ’Ali, not to recede from 
his conquests, but “to limit his ambition at the point at 
which he has admitted it should be satisfied, and to 
prefer an arrangement with the Porte to the continua- 
tion of the war . . . ”. 1 

The attitude of the English ministry was not dis- 
similar. Barker, the consul-general, who was probably 
influenced by consular opinion in Syria where he had 
served and where the advantages of the corrupt 
Turkish regime were fully understood, disapproved of 
Ibrahim’s victories, refused to pay Muhammad ’Ali a 
visit of congratulation on the fall of Acre, 2 and de- 
lighted after the promulgation of the viceroy’s removal 
from office by the Porte to refer to him as “the ex- 
viceroy” or “the rebel”. But this by no means repre- 
sented the views of the London Foreign Office. Pal- 
merston, who had accepted the Secretary’s seals at the 
close of 1830, not only gave Barker a sharp rebuke for 
thus venturing to take for granted the decision of His 
Majesty’s government, 3 but shortly afterwards replaced 

1 Douin, Mission du Baron de Boislecomte, p. iii. 

2 Barker, June 13, 1832 (F.O. 78-214). 

3 To Barker, October 3, 1832 (ibid.). 


him by Colonel Patrick Campbell , 1 by far the ablest 
and most understanding of all the British representatives 
in Egypt in Muhammad ’Ali’s time. 

In order to render his aggrandisement less alarming 
in western eyes, the viceroy, as he still was, however 
much his sovereign might disclaim him, proceeded to 
develop a somewhat paradoxical thesis which he had 
already pressed upon Great Britain. That was the view 
that he remained at heart a humble servant of the 
Turkish empire, if not of the Turkish Sultan, that he 
was acting only for the advantage and glory of the 
Sublime Porte, that he had no views of independence, 
and that he was conquering Syria merely in order to 
consolidate Turkish power . 2 But, since Sultan Mahmud 
had been proved by experience incapable of guiding 
the Turks anywhere except to defeat, and since the 
Divan was incurably hostile to the only man — that is, 
Muhammad ’Ali himself— who could save the empire 
from ruin, it had become his clear duty as a patriotic 
and loyal Turk to dethrone Mahmud and set his young 
son, Abdul Majid, up in his stead with such a Divan as 
would ensure the wise conduct of affairs . 3 A tentative 
move was made in August and September to abolish 
the only visible sign in Egypt of Mahmud’s sovereignty. 
Under cover of the fact that the Turkish coinage had 
long undergone a steady course of debasement and was 
becoming worse, its currency was prohibited through- 
out Egypt, in order to prevent it from replacing the 
European and Egyptian coins in use in the country . 4 
This had in reality no connection with Gresham’s 
famous law. It was an ingenious attempt under the 
guise of an economic reform to proclaim to the people 

1 To Campbell, January 7, 1833 (F.O. 78-226). 

2 Barker, June 25, 1832 (F.O. 78-214). 

3 The same, August 12, 1832 {ibid.). 

4 The same, September 19, 1832 {ibid.). 




1 14 

of Egypt that they were no longer governed in the 
name of Sultan Mahmud. 

Meanwhile discussions had been continuously pro- 
ceeding between Alexandria and Constantinople. The 
Sultan had sent commissioners at the end of 1831, who 
had been received with much graciousness; but the 
discussions were prolonged over the next two months, 1 
and ended in nothing more lasting than the fumes of 
the chibouques they puffed day after day in the 
viceroy’s palace. Then followed an indirect negotiation 
conducted through the Turkish Capitan Pasha. In 
September Muhammad ’Ali told Barker that as he had 
had no satisfactory answer, there was nothing left for 
him but to march on Constantinople and that he had 
secret news from that place that “there is now nothing 
to prevent my doing so”. 2 But in November he was 
still ready to negotiate with any agent whom the Porte 
chose to send to Alexandria, 3 and directing Ibrahim 
not to proclaim the abolition of the Sultan’s authority 
in Syria unless he could first obtain fatwas from the 
local religious leaders declaring Mahmud deposed as 
unfit to rule. 4 

But under cover of these palavers, the Sultan was 
preparing a final effort to drive the rebel’s forces out of 
Syria. Indeed his negotiations were apparently de- 
signed to throw the enemy off his guard and divert him 
from either advancing on Constantinople or increasing 
his forces until the Turkish preparations had been 

Ibrahim with his army had advanced as far north- 
ward as Koniah — the ancient Iconium — where he had 
been detained by his father’s fear of provoking inter- 

1 Barker to Mandeville, January 2, and to Hotham, February 
28, 1832 (F.O. 78-213). 

2 Barker, September 20, 1832 (F.O. 78-214). 

8 Sabry, op . cit. p. 208. 4 Idem , p. 212. 


vention. At the close of the year 1832 the Grand Vazir 
himself, Rashid Muhammad Pasha, advanced against 
him. Rashid had a numerous army, stiffened with 
large levies from Bosnia and Albania. On December 
2 1 the armies met close to Koniah. The Turk cavalry 
early left the field; the infantry were drawn on into an 
attack in the course of which they suddenly found 
themselves fired on from either flank. They were 
utterly routed; the Grand Vazir was taken prisoner; 
and the way to Constantinople lay open and defence- 
less. Ibrahim resolved to advance at once, in the hope 
of confronting Europe with the accomplished fact of 
the Sultan’s overthrow. But at Kutahia he received a 
letter from his father directing him to halt, wherever 
he might be. 

This decision was brought about by the very inter- 
vention that Muhammad ’Ali had always feared. On 
January 12, 1833, rumours had already reached Alex- 
andria that the Turks had accepted a Russian alliance. 1 
The news in fact was premature, for though the 
Emperor of Russia had offered the Sultan armed 
assistance against Muhammad ’Ali, the offer had not 
yet been accepted; but almost immediately afterwards 
a Russian officer, Lieutenant-General Muravief, had 
reached Constantinople with orders to proceed at once 
to Alexandria and require the viceroy to desist from his 
attacks on Turkey. He reached Alexandria on January 
13. Early next morning he had a short audience. He 
presented no written document to the viceroy; and it 
was announced that he had come as a mediator of 
peace. But it was generally believed that his mission 
was to call upon Muhammad ’Ali to retire from Cara- 
mania and Syria, to give up his fleet to the Sultan, and 
to reduce his army to 20,000 men. Two days later, and 
again on the 18th, he had two more conversations with 
1 Barker, January 17, 1833 (F.O. 78-231). 

8 -a 


the pasha, prolonged and secret. Muhammad ’Ali 
gave way, promised to submit to the Sultan, and, as a 
pledge of good faith, to suspend hostilities . 1 

The Turkish Divan had naturally looked for help in 
its distressing circumstances to England, its traditional 
ally, rather than to Russia, its traditional foe. A special 
envoy had been sent to London to seek the aid of 
English men-of-war. But Palmerston was not prepared 
for such definite action. He refused. This rebuff 
determined Mahmud, sorely against his will, to make 
terms with the rebellious viceroy. On January 21 
therefore Khalil, the Capitan Pasha, reached Alex- 
andria, with offers of peace. 

The meeting of the two pashas was ceremonious in 
the extreme. Khalil was assisted up the staircase of the 
Ras-al-tin Palace by two of Muhammad ’Ali’s principal 
officers. The viceroy himself advanced half-way down to 
meet him, would not allow the other to kiss his hand, but 
embraced him and kissed him on the cheek. They pro- 
ceeded to the audience chamber hand in hand, Khalil’s 
disengaged arm clasping the viceroy’s portly waist. 
When they took their seats, Khalil respectfully doubled 
up his legs under him. So much etiquette was a natural 
prelude to prolonged, tedious, insincere discussions. 

What terms of peace could Muhammad ’Ali hope to 
secure? Ibrahim from his army headquarters at 
Kutahia suggested a list of demands. In the first place 
he set independence — “a vital question for us, domi- 
nating all the rest”. Then Anatolia and Cilicia, for the 
sake of the timber they might supply — an article which 
Egypt must obtain from abroad, if the fleet was to be 
maintained. Lastly Cyprus, as a base for the fleet. 
Baghdad he thought of small moment; it was too 
distant and too poor . 2 From the Egyptian point of 

1 Barker, January 17 and 19, 1833 (F.O. 78-231). Sabry, op. 
cit. p. 223. 2 Quoted ap. Sabry, op. cit. pp. 227-8. 


view all were most desirable. But the only claim which 
could be laid to any of them was the claim of the 
conqueror. Such claims can be made good only by 
superior force; and there could be no reason why 
Europe should acquiesce in such demands if they 
seemed politically undesirable. 

An instructive contrast to Ibrahim’s view is afforded 
by the instructions which at almost the same time 
Palmerston was drawing up for Colonel Campbell. 
“His Majesty’s Government”, he wrote, “attach great 
importance to the maintenance of the integrity of the 
Ottoman Empire, considering that state to be a 
material element in the general balance of power in 
Europe, and they are of opinion that any considerable 
encroachment upon the Asiatic territories of the 
Sultan, and any consequent defalcation from the re- 
sources which he might bring to bear for the defence of 
his European dominions, must operate in a correspond- 
ing degree upon his relative position with respect to 
neighbouring powers, and must thereby have injurious 
bearings upon the general interests of Europe. His 
Majesty’s Government therefore deem it of importance 
to prevent not only a dissolution but even a partial 
dismemberment of the Turkish Empire.” It was 
obviously impossible to re-establish the former state of 
things, but in the circumstances the best solution was to 
assign Syria to Muhammad ’Ali on such terms of 
tribute and military aid as would leave the Porte’s 
revenues and resources undiminished . 1 

And the fact was that Ibrahim’s impatience, his 
reliance on the sword despite his experience in the 
Morea, had turned the scales decidedly against his own 
and his father’s ambitions. The news of his decision to 
advance upon Constantinople, arriving just after the 
despatch of Khalil to Alexandria, had filled the Divan 
1 To Campbell, February 4, 1833 (F.O. 78-226). 


with wild, but not exaggerated apprehension. There 
were no troops but the broken fragments of the late 
Grand Vazir Rashid’s army to bar the way. All had 
gone on that most unlucky throw at Koniah. Ibrahim’s 
approach would arouse every smouldering ember of 
discontent; Muhammad ’Ali’s agents would fan them 
into a blaze. The empire would collapse, the Sultan 
would fall, the ministers would lose office and perhaps 
life. In their panic they turned to Russia who already 
had offered military aid. They begged her to send men- 
of-war and at least 20,000 men to save Constantinople. 
And Russia eagerly agreed. Even when Muravief 
returned from Alexandria with the news that Ibrahim’s 
march had been stayed, and when both English and 
French representatives assured the Porte that military 
aid was no longer needed, the Divan still refused to 
withdraw its request, and a Russian force encamped on 
the Asiatic side of the Bosphorus. Ibrahim in fact had 
committed a ruinous blunder. For it was not merely 
the Russians who were roused. The western powers, 
who had been mere spectators in the earlier phases of 
the war, were now drawn into inevitable action. It had 
become necessary to end the Syrian war, to free Con- 
stantinople from fear, to get rid of the Russians as soon 
as possible, to protect the Turks, if fate permitted, from 
the consequences of their own misconduct, lest the 
partition of their empire should set all Europe by the 
ears. In vain Muhammad ’Ali sought to retrieve 
Ibrahim’s slip by urging once more his pet theory of 
regenerating the empire by way of rebellion. He had 
never dreamt of independent power, he assured Camp- 
bell, and, by supporting him, France and England 
would be supporting the Sultan in the most effectual 
way possible . 1 “Cool and clear-sighted observation”, 
ran a memorandum which he gave to Campbell, 
1 Campbell, March 31, 1833 (F.O. 78-227). 


“shows that the Turkish government is worn out on 
every side, with its foundations cracking beneath it, 
having exhausted its physical and moral resources, 
abandoned and despised by the nation , 1 discredited 
and distrusted at Constantinople itself, incapable of 
defending either itself or the nation; it has in short 
abandoned itself to its fate and lies a helpless prey in the 
Russian talons. . . . ” 2 But while the pasha’s assertions 
corresponded not unfairly with the facts as they 
appeared to contemporary Europe, western statesmen 
were indisposed to admit that Muhammad ’Ali was the 
sole possible regenerator of the empire. The western 
nations themselves might come to the aid of the Sultan, 
and the maintenance of the Sultan’s position as little 
weakened as might be seemed to them to promise better 
chances of gently edging out the Russians from their 
extraordinary and unprecedented attitude of champions 
of the Turk, than all the forces which Ibrahim had at 
his disposal. 

Another argument which the pasha strove to intro- 
duce was the principle (as we should call it nowadays) 
of self-determination. The case offers a delightful 
example of the false facility with which the political 
principles of the western world may be employed to 
mask action essentially different in character. Muham- 
mad ’Ali claimed to be acting on behalf and in the name 
of the “nation” of Islam. He had urged on Ibrahim 
the need of obtaining from the learned men of Syria 
declarations that Sultan Mahmud was or should be 
deposed as incapable of rule. In reply Ibrahim had 
pointed out to him the patent absurdity of expecting 
the ulema of Damascus to renounce the Sultan’s 
sovereignty before another had laid claim to it and 
proved his right by force. And a further difficulty lay 

1 Scilicet , the world of Islam. 

2 Campbell, ut supra. 


in the fact that the western nations had consular agents 
all over Syria, so that it would be hard to obtain de- 
clarations without knowledge leaking out of the means 
by which alone such declarations could be secured. 
But what could hardly be got in Syria without certain 
exposure of the coercion or bribes employed might of 
course be asserted of more distant regions where there 
were no consular agents. A declaration was produced 
purporting to have been made by a group of Kurds 
bordering on the Black Sea, renouncing their allegiance 
to the Sultan and offering their obedience to the Pasha 
of Cairo. It was not a little singular, as the French 
consul-general remarked, that such a declaration should 
come from a province which, at the moment at all 
events, Muhammad ’Ali could not possibly protect 
against the Sultan’s agents, and that the persons 
bearing it should have been able to travel by Angora 
undisturbed . 1 

In European eyes, at all events, such arguments 
carried little conviction. The pasha might claim to be 
regarded with the same sympathy as Europe had 
accorded to the Belgians and the Greeks, but even his 
persuasive tongue could not disguise the fact that he 
was fighting for his own hand. He did not represent any 
nation struggling to be free. His military superiority 
over the Turks could give no moral claim to special 
consideration. His only moral claim — if so it can be 
called — lay in the superior order, justice and regularity 
which it might be expected he would introduce into his 
new conquests as he had introduced them into Egypt. 
But even then, since his methods would certainly be 
those of oriental administration, western statesmen 
would still find opportunities for criticism and doubt. 

1 Mimaut, December 25, 1832, quoted ap. Sabry, op . cit. pp. 
230-1. M. Sabry, I think, misapprehends the significance of the 
concluding part of the despatch. 


Political expediency was therefore the only standpoint 
from which the matter could reasonably be discussed at 
Paris and London. 

France and England were at least thoroughly agreed 
on the necessity of eliminating the suddenly arisen 
Russian influence at Constantinople, and on the need of 
arresting Ibrahim’s advance in order to quiet the 
extraordinary alarm which had seized upon the Porte. 
They therefore called upon Muhammad ’Ali to with- 
draw from Asia Minor, and even threatened to blockade 
Alexandria unless he complied with their demand. 1 
But while Palmerston was entirely indisposed to admit 
any alteration in the pasha’s theoretical position of 
vassalage to the Sultan, the French were inclined to toy 
with the idea of recognising him some day as virtually 
independent, like the Barbary deys, and to use this as a 
means of inducing him to accept unpalatable terms 
with a minimum of ill-feeling. They even sent a minister 
plenipotentiary to Alexandria. This diplomatic blunder 
provoked the Austrian representative to enquire to 
what court the plenipotentiary was accredited, and the 
French consul-general was obliged to declare that the 
minister claimed no rank above that of envoy on a 
special mission. 2 

While the western powers were thus seeking to cajole 
or to alarm the viceroy into withdrawing his troops, the 
Porte suddenly yielded to Muhammad ’Ali’s demands 
so far as to grant him Crete and the four Syrian pasha- 
liqs, withholding only the district of Adana. The news 
arrived on April 1 6 . The messenger was received in the 
presence of the English and French consuls-general. 
On the delivery of his message, “the pasha started up 
with tears of joy in his eyes, and, laying aside anything 
like Turkish gravity, burst into a sort of hysterick 

1 Campbell, April 9, 1833 (F.O. 78-227). 

2 The same, May 3, 1833 {ibid.). 


laugh”. 1 He doubtless assumed that this was a sign 
that the Porte was weakening and would in a little 
concede Adana as well. But both England, France 
and Austria continued to press on him the need of 
yielding. He was indeed brought to declare “that he 
was ready to give up all claims to the government of 
Adana, and to promise besides to all the great powers 
to remain for ever the faithful vassal of the Porte and 
not to disturb his sovereign in any way, provided the 
Porte would declare on her part to the representatives 
of those powers that she (the Porte) would never 
attempt to withdraw the rights which she had granted 
to him”. 2 A few days later, in conversation with the 
French “plenipotentiary”, he observed in a like spirit: 
“I am a peaceable man, who aims at nothing else but 
to consecrate the rest of his days to the prosperity of the 
countries that I have to govern. They ask me the proof 
of this disposition. I give it by begging of Europe to 
guarantee Turkey from all aggression on my part, and 
at the same time to guarantee me from all aggression on 
her part”. 3 

These negotiations however proved otiose, except in 
giving the pasha an opportunity of developing his 
views, for on May 3 the Porte had already given way on 
Adana as well. Thus the various points had now all 
been settled, except the question of the tribute to be 
paid the Porte for the ceded provinces. That was agreed 
in the following September, on the basis of 30,000 
purses a year for the governments of Egypt, Syria, 
Adana, and Tarsus. 4 

The Syrian war thus ended to the satisfaction of no 
one. The Sultan had suffered the vexation of defeat by 

1 Campbell, April 17, 1833 (F.O. 78-227). 

2 The same, May 9 (P.S. May xo), 1833 (ibid.). 

8 The same, May 13, 1833 (ibid.). 

4 The same, September 13, 1833 (F.O. 78-228). 


a contumacious pasha; Muhammad ’Ali had secured 
neither independent status nor a controlling influence 
at the Porte ; the western powers were annoyed at the 
opening which Ibrahim’s victories had offered to the 
Russians; while the Russians were disappointed at 
having been unable to entrench themselves more 
securely at Constantinople. However, they did not 
withdraw until they had obtained by the secret clause 
of the Treaty of Unkiar Skelessi, which was signed on 
July 8, the right of closing the Dardanelles against any 
foreign man-of-war. Here is perhaps to be found the 
source of that unconquerable distrust with which 
Palmerston came to regard the policy of Muhammad 
’Ali. Even before the conclusion of the treaty, Palmer- 
ston, though not actively hostile, had been unfavourable 
to Muhammad ’Ali’s plans. “His real design”, he had 
written, “is to establish an Arabian kingdom, includ- 
ing all the countries of which Arabic is the language. 
There might be no harm in such a thing in itself; but as 
it would necessarily imply the dismemberment of 
Turkey, we could not agree to it. Besides Turkey is as 
good an occupier of the road to India as an active 
Arabian sovereign would be.” 1 Such an attitude was 
natural towards the schemes of one whose ambition 
had just raised the most difficult question of European 
politics in a most acute form. That Anglo-Egyptian co- 
operation, which the viceroy had hoped to secure, as a 
consequence of Turkish feebleness, or of European 
jealousies, had quite definitely become less probable. 
A political principle — the assertion of national indepen- 
dence or the substitution of political liberty for des- 
potism — might form a reasonable ground for political 
disturbances and could at least be reckoned on to evoke 
considerable popular sympathy. But the mere sub- 
stitution of an efficient for an incompetent autocrat 
1 Bulwer, Life of Palmerston, n, 144-5. 


aroused no particular sentiment in the whig or liberal 
breast. The process of reform which Muhammad ’Ali 
had undertaken, the beneficent consequences of a 
stern, regular, and highly despotic government, the 
infusion into heterogeneous peoples of that common 
consciousness without which nationality is impossible, 
the civilising influences which his administration was 
gradually spreading — all these were obscured by stories 
of the severity of his conscription, the ferocity of his 
punishments, the oppressiveness of his monopolies. 
Palmerston must not be too severely blamed if he failed 
to grasp the significance of Muhammad ’Ali’s career. 
In his eyes Muhammad ’Ali was above all the man 
whose inordinate ambition had almost established the 
Russians in a dominant position on the Bosphorus. 

Chapter V 


Palmerston had already, as has been seen, attributed to 
Muhammad ’Ali the project of establishing “an 
Arabian kingdom”, embracing all the countries in 
which Arabic was the common form of speech. Such an 
ambition came very naturally to the viceroy. The con- 
quest of Syria, following on the conquests of Egypt, of 
the Hijaz, and of the Sudan, left him with compara- 
tively few further acquisitions to make in order to 
bring that plan within sight of realisation. There 
remained only ’Iraq and the Persian Gulf, together 
with southern Arabia, in order to complete such a 
territorial expansion. From a financial point of view, 
none offered any particular advantages except perhaps 
the pearl fisheries of Bahrein, while in a military sense 
all were inhabited by nomadic or quasi-nomadic tribes, 
who would bitterly resent regular government, es- 
pecially when accompanied by the imposition of taxa- 
tion or the establishment of conscription. But while 
these areas might be of small value in themselves, their 
occupation did in fact promise certain advantages. 
’Iraq would bring the viceroy’s territories into touch 
with Persia and through Persia with central Asia. 
Southern Arabia would render him master of those two 
great inlets of the sea, the Red Sea and the Persian 
Gulf, at all times of great strategic importance, and 
perhaps enable him to allow or deny their use to the 
British squadrons in the East. Such an expansion 
therefore, though adding little to Muhammad ’Ali’s 
material resources, might greatly increase his political 
weight and influence. 


He was convinced, and with good reason, that southern 
Arabia could offer no sustained resistance to any well- 
organised force, and that Baghdad in all probability 
would offer none at all. The general condition of the 
pashaliq is described as wretched in the extreme. “In 
their misery”, wrote Colonel Taylor, the East India 
Company’s agent, “the people look to Ibrahim .” 1 
Merchants at Baghdad could see no limits to the 
rapacity of the Turkish government save its fear of the 
despatch of troops from India, and they deplored the 
decision of Palmerston to prevent the addition of the 
province to what they had already begun to call “the 
Egyptian caliphat ”. 2 

Indeed, could Muhammad ’Ali have asserted his 
independence, the Egyptian caliphate would certainly 
have been revived. He controlled and protected the 
Hijaz; and, whatever subtleties theologians might have 
spun around the caliph’s holy office, the popular view 
was that the Sultan’s caliphate could not survive the 
disappearance of his nominal authority at Mecca and 
Medina. As Ibrahim wrote to his father, the Sultan 
could no longer be prayed for in the mosques as the 
Servant of the Holy Places . 3 Even before the Syrian 
war it had been common talk in Egypt that the sharif 
of Mecca was about to publish a declaration “ that he 
who is the possessor and defender of the Kaba is the 
true head of the Mahometan church ”. 4 

But besides the possession of the Hijaz, Muhammad 
’Ali controlled another source of influence. Mecca 
might be the spiritual metropolis of Islam, but it was 
not a centre of Islamic culture or learning. It possessed 

1 Taylor to Campbell, November 6, 1833 (F.O. 78-288). 

2 Tod to the same, November 27, 1833 {ibid.). 

3 Sabry, op. cit. p. 281. Cf. Ponsonby, No. 305. November 7, 
1839 (F.O. 78-360). 

4 Barker to S. Canning, February 23, 1832 (F.O. 78-213). 


no considerable school. It had no great libraries. It 
had neither bookshop nor bookbinder. True, lectures 
were delivered in the great mosque, but they were 
never delivered by any of the great literati of Islam, and 
were attended only by a few ignorant Indians, Malays, 
and negroes . 1 Cairo and Damascus were the true 
homes of Muslim learning at this time; and now 
Muhammad ’Ali governed both. This gave him great 
weight in the Muslim world; and could he have de- 
veloped those cities into conscious centres of Arab, as 
opposed to Muslim, culture, and stood forth as the 
champion of Arab against Turk, he might perhaps have 
created in his territories some stronger bond of union 
than subjection to a common master. He was indeed 
reproached for his failure to achieve this idea . 2 But 
such criticism seems to ignore essential factors in the 
situation as it was. Islam is a faith that has never en- 
couraged the growth of nationality. Its universal 
character has toned down, rather than accentuated, 
racial and cultural differences that might have hardened 
into national qualities ; and it is noteworthy that even 
a century afterwards, even under the prolonged in- 
fluence of western education and ideas, nationalist 
rulers have been rather hindered than assisted by its 
unlimited, catholic claims. Nor was that all. The only 
common factors in the Arab world were unity of 
language and unity of subjection. The Syrian and the 
Egyptian, the nomad and the fellah, the learned and 
the populace, were too much divided by custom, by 
ideas, by tradition, to be at all willing to recognise any- 
thing common but religion. So that Muhammad ’Ali 
found himself obliged to pose as the champion of the 
Muslim “nation”, not of an Arab nation, which indeed 
he could not even imagine. As Ibrahim was to find, the 

1 Burckhardt, Arabia, 1, 389-391. 

a Puckler-Muskau, op. cit. 1, 210. 


differences between Syrian and Egyptian were far too 
strong and deep for any assimilation to be possible; 
and the attraction for Muhammad ’Ali of the territories 
occupied by the speakers of the Arabic tongue lay 
in his just sense of their strategic importance rather 
than in any anticipation of conditions which in his 
day were scarcely conceivable. The idea of an Arab 
nationality has been begotten and brought forth only 
in our own day, under the pressure of western influence, 
of the spread of education, of a popular press, and 
above all of an extraordinary development of com- 

His idea, therefore, was not to create an Arab unit 
within the circle of Islam, but to become, and be ac- 
claimed by all, the foremost leader of Islam itself. This, 
however, would involve either the overthrow of the 
Sultan and the dismemberment of his dominions, or the 
overthrow of the Divan at Constantinople and the 
substitution of Muhammad ’Ali’s for Khusrau Pasha’s 
influence. His position, always anomalous, had be- 
come yet more evidently so after the Syrian war and 
Ibrahim’s victorious campaign. “He is ”, wrote Camp- 
bell with truth, “ de jure a vassal and de facto he is inde- 
pendent; and although he makes professions of his 
being a vassal and subject of the Sultan, it is in such a 
manner as to lead me to think that he would not wish 
anyone else to suppose so .” 1 Both French newspapers 
and official French communications nourished in him 
the belief that a declaration of his independence would 
meet with much sympathy and support. The evident 
(and natural) ill-will displayed towards him by the 
Sultan and his ministers pressed him strongly in the 
same direction. “These late menaces and hostile demon- 
strations of the Porte”, Campbell wrote again a week 
later, “will doubtless tend to confirm Mehemet ’Ali in 
1 Campbell, August 15, 1834 (F.O. 78-246). 


his desire of independence and in the object which I 
feel almost certain that he contemplates eventually, of 
forming an Arab caliphat. . . . He is naturally very 
ambitious of power and glory, and unlike the generality 
of Mussulmans he is actuated by the strong desire of 
handing down his name to posterity in the page of 
history. Success has ever attended him.. . .” x 

The Sultan’s policy in calling in Russian aid had 
intensified at once Muhammad ’ Ali’s scorn and his dislike 
of the conduct of affairs at Constantinople. It had of 
course introduced an unexpected element into the 
game. It was a thrust for which he commanded no 
suitable parry. Hence one strong reason why he should 
condemn it openly and bitterly. But, quite apart from 
that, it violated Muslim sentiment and threatened 
Muslim unity. It had alienated the affection of the 
whole people, he declared, and the newly arrived qazi of 
Cairo, whose appointment was one of the few relics of 
Turkish dominion in Egypt, said — it was of course 
worth his while to conciliate the pasha’s good opinion — 
that he had been assured by many of the chief people at 
Constantinople that they looked to Muhammad ’Ali 
“as the chief support of the empire in the event of any 
future war with Russia ”. 2 If the Turko-Russian 
alliance could be countered by an Anglo-Egyptian 
understanding, all the dreams which the viceroy had 
cherished in his heart for years might yet come true. 

Accordingly an interesting memorandum was given 
to the English consul-general for transmission to 
London. It was the viceroy’s first object, it asserted, to 
root Russian influence out of Turkey, and to organise 
such a force as would compel the Russians to respect the 
independence, not only of Turkey, but of Persia as well. 
“The viceroy’s desire of possessing Syria was inspired 

1 Campbell to Ponsonby, August 21, 1834 (F.O. 78-246). 

2 Campbell, June 25, 1833 (F.O. 78-227). 




by that single purpose; and after the battle of Koniah 
he had hoped to bring about such a change in the 
government and policy at Constantinople as with the 
help of England and France would have speedily dis- 
concerted the designs of Russia.” He would (the 
memorandum continued) soon have an army of 150,000 
men ready to co-operate with England in the glorious 
task of delivering Turkey and Persia from the Russian 
yoke. Meanwhile he demanded of British equity and 
justice whether he would not be justified in declaring at 
once his independence, as he had resolved on doing 
should the Porte’s hostility continue. 1 

At the moment the English agents in the East were 
much disposed to assent to these views. In 1833 
Ponsonby had written to Campbell, “ If Russia should 
look to selfish ends, it is to be hoped that the strength 
of Mehemet Ali may be found efficient where it seems 
certain it must be his interest to exert it, viz. in driving 
from Asia and the Ottoman territory a power, which, if 
allowed to spread its roots there, will easily ere long be 
great enough to destroy the new-born energies of his 
Egyptian and Arab population”. 2 Campbell himself 
in the next year thought “that so far as regards the 
resistance to Russian encroachments and aggrandize- 
ment on the side of Asia, perhaps the establishment of 
an Arab caliphat under Mehemet Ali would be a better 
barrier and more likely to afford effectual opposition to 
Russia than the Porte could now ever be expected to 
offer, and in case of need Mehemet Ali could give great 
assistance to Persia (supposing him to rule over Bagh- 
dad etc.) in any struggle of Persia against Russia”. 3 

To some extent these views may have coincided with 
Palmerston’s dislike of Russian policy and aims. He 

1 Boghoz Bey to Campbell, September 3, 1834 (F.O. 78-246). 

2 Ponsonby to Campbell, May 24, 1833 (F.O. 78-227). 

3 Campbell to Ponsonby, August 21, 1834 (F.O. 78-246). 


regarded her as the only European power with which 
we were likely seriously to quarrel, and he complained 
of her system of universal aggression, inspired alike by 
the personal character of the emperor and by the per- 
manent character of her government . 1 He was, too, 
receiving at this time information from persons in no 
way under the command of Muhammad ’Ali’s power 
or the spell of his persuasion that the Russians were 
seeking to strengthen themselves in the critical area of 
’Iraq. Once their influence was established at Baghdad, 
our agent wrote, “its central position and its navigable 
streams and natural resources will afford the highest 
advantages to future advances ... or the establishment 
and continuation of intrigue more fatal than war ”. 2 
Might not the threat of Russian intrigue, the fear of 
Russian advances through Persia towards India, secure 
at last for the viceroy that British recognition and aid 
which neither the evacuation of the Morea nor the 
project of alliance with France had been able to 
obtain? Had not the British government in India 
sought the alliance of the Sikhs, the Afghans, and the 
Persians when it feared that Napoleon might advance 
overland upon India? 

But these calculations left out of account the position 
of Great Britain and the character of her foreign 
minister. At this time she was fully conscious, it may 
have been over-conscious, of her power and her re- 
sponsibilities. For five generations she had lost one war 
only, when her arm had been weakened by the know- 
ledge that those she fought were of her own household. 
Her last war had been not only the fiercest but the 
most triumphant of them all. Was she then likely to 
change the whole basis on which her European policy 

1 Palmerston to W. Temple, December 3, 1833 (Bulwer, op. cit. 
n, 176). 

8 Taylor to the India Board, March 14, 1834 (F.O. 78-245). 


was built in order to buy an alliance against a possible 
foe who had never in an offensive war beaten anyone 
but Turks and Persians? Nor was Palmerston ever the 
man to seek the prop of a foreign alliance instead of 
developing the power and resources of his own country. 
His limitation lay in bluntness of intellect, not in any 
defect of courage; and he was bent on hindering the 
progress of Russia by other means than those which 
Muhammad ’Ali was suggesting. 

He therefore returned the most uncompromising 
answer. Campbell was ordered to express regret and 
surprise at proposals so at variance with Muhammad 
’Ali’s former professions and incompatible with the 
honour and good faith of the British government. 
Muhammad ’Ali desired “that Great Britain should 
either consent to an attack to be made by the pasha 
upon the Sultan or she should sanction an attempt on 
the part of the former to throw off his allegiance to the 
latter and to declare himself the independent ruler of 
the provinces which he at present administers as the 
vicegerent of his sovereign”. How could we sanction 
rebellion and usurpation directly against a crowned 
head in alliance with His Majesty ? 1 

In this language there was doubtless an element of 
absurdity and perhaps of falsehood. Palmerston was 
writing of Muhammad ’Ali’s attitude towards the 
Sultan much as though the relations of that derelict 
monarch and his ministers were those familiar in the 
West. The Secretary of State indeed approached the 
matter as he would have expected the United States to 
treat similar offers from the Governor-General of 
Canada, or France the like suggestions from the 
Governor-General of India. Acceptance or even en- 
couragement could be justified only by a state of war or 
at the least a state in which war was confidently antici- 

1 Despatch to Campbell, October 26, 1834 (F.O. 78-244). 


pated, while the minister capable of seeking foreign 
aid against his sovereign was clearly guilty of the 
blackest treason. But such views evidently took for 
granted conditions which were notoriously non-ex- 
istent. The Governor-General of Canada could rest 
assured that the success of his administration would not 
expose him to the hatred and vengeance of his king; the 
Governor-General of India could rely on the prime 
minister’s not seeking his disgrace and execution. The 
political ideas of the West were therefore being applied 
to a world in which they were imperfectly apprehended 
and completely ignored. 

But the recognition of this does not perhaps invali- 
date Palmerston’s general attitude. Turkey had be- 
come a part of the state system of Europe. Alliance 
with her carried the same obligations as alliances with 
other powers — obligations which could not really be 
affected by the chaotic state of her internal affairs. So 
much was completely undeniable. And in this case the 
guidance of political principle was strengthened by 
considerations of political expediency. It would have 
been entirely permissible to denounce our long-standing 
alliance with the Sultan and then to support Muham- 
mad ’Ali in his projects against the Turkish empire and 
the Ottoman caliphate. But the art of foreign policy 
consists in pursuing national interests within the limits 
imposed by a due observance of political principle. The 
latter would have been violated by any secret count- 
enance afforded to the viceroy of Egypt; the former 
would have been threatened by an open union with 
him. The withdrawal of our support from the Sultan 
would have meant an instant scramble for the inheri- 
tance of his empire, — a contingency which we could 
not view with composure, for we should scarcely have 
been benefited by seeing the Adriatic become an 
Austrian lake, or Constantinople a Russian port. What 


advantage could Muhammad ’Ali offer to make it 
worth while to turn Europe upside down, or why 
should we gratuitously aid the ruler of Egypt to extend 
his power by military conquest into new areas to which 
he could not allege the least shadow of a rightful claim? 
Out of these considerations emerged a policy of sup- 
porting the maintenance of Muhammad ’Ali’s authority 
in the territories already ruled by him but of opposing 
consistently any further expansion of his power. 
Palmerston sensibly preferred to strengthen our own 
position on the newly developing routes to India rather 
than help to build up a new state in the hope that some 
day that new state would throw in its lot with us in a 
possible war with Russia. 

As matters stood, the two possible overland routes to 
India — by the Euphrates and by Suez — had by no 
action of our own fallen from under the control of a 
single political authority. The rise of Muhammad ’Ali’s 
power in Egypt had given him the Suez route, while 
the Euphrates valley still remained under the govern- 
ment, however nominal, of the Sultan. It would in- 
deed have been an act of great folly to have assisted, 
without any moral or political obligation, in placing 
both under Muhammad ’Ali’s command at the very 
moment that they were coming into high political im- 
portance . 1 The primary cause of this development had 
been the application of steam-power to shipping. So 
long as the Red Sea route had been closed during 
several months of the year by the prevailing winds, and 
so long as the passage up the Euphrates could be ac- 
complished only by the tedious method of tracking, 
those ways to the East, while of undoubted military 
importance, could not rival the long sea voyage round 
the Cape. But before the close of the Napoleonic wars 

1 Cf. Palmerston to W. Temple, March 21, 1833 (Bulwer, op. 
cit. 11, 144). 



steamboats were in use on English rivers and canals, and 
a few years later they were already employed in channel 
crossings. By 1 820 men were anticipating their introduc- 
tion on to the great ocean routes. But here progress was 
more difficult. The early marine steam engine was feeble 
and wasteful. It consumed enormous quantities of coal, 
and so was closely tied to the coastline where alone it 
could renew its stores of fuel. The clumsy paddle-boxes 
were liable to be torn away by high seas, and the engines 
themselves had to be stopped frequently for cleaning and 
repair. In these circumstances their earliest use was evi- 
dently destined to be limited to regions offering a con- 
tinuous chain of ports — the Channel, the Mediterranean, 
the Red Sea, or the Persian Gulf. 

The importance of these possible developments was 
rapidly grasped in India. In 1823 the Calcutta mer- 
chants formed a committee to investigate the matter. 
This led to the voyage of the Enterprize , which spent 1 13 
days half-steaming, half-sailing, from London to 
Calcutta round the Cape. This comparative failure 
demonstrated the disadvantages of the long sea route 
for these primitive steamers and turned public attention 
to the more appropriate route by Suez and the Red 
Sea. This had been advocated by Mountstuart Elph- 
instone at the time when the Calcutta committee had 
begun its campaign. His successor in the government 
of Bombay, Sir John Malcolm, took up the idea with 
enthusiasm. In 1829 he attempted to send the Enterprize 
from Bombay to Suez. He had a new vessel, the Hugh 
Lindsay , specially built, and in 1830 she accomplished 
the first steam voyage up the Red Sea. Although at 
this moment the East India Company could not be 
induced to complete the plan by chartering steamers to 
Alexandria and back, to meet the mails and passengers 
from Suez, other experimental voyages were made, 
admiralty packets sailed from Malta to Alexandria, a 


select committee was appointed to report on the whole 
question of steam communication with India, and 
merchants began largely to use the Suez route for 
letters even before any regular service had been estab- 
lished. 1 Thomas Waghorn, who had been active in 
promoting the scheme, established himself at Alex- 
andria and began to act as an agent for forwarding 
correspondence, although the East India Company 
still refused to make use of the route. Auckland, the 
Governor-General of India, in 1 836 drew a lively picture 
of the condition of affairs. “The merchant”, he writes 
to Hobhouse, “receives at the India House bills upon 
our treasury at so much after sight, which at the 
Court’s rate of travelling [i.e. via the Cape] should be 
four or five months after date. He sends them to 
Alexandria.. . .The indefatigable Waghorn gets into a 
boggoora [a dhow] with the letter-bags, finds his way to 
Mocha, puts them on board a trader, and they are at 
Calcutta a little more than two months after leaving 
England. The merchants here have all their letters; 
their bills for 20 lacs are paid at our treasury to their 
profit and our loss; private correspondence flourishes; 
newspapers pour in ; and the government upon public 
business has not a single line from any authority. As 
far as I, as governor-general, am concerned, I would 
rather have the communication by the Cape of Good 
Hope than the Red Sea and by Cape Horn than either ; 
but if the short path is open, it is strange and incon- 
sistent that all should use it except those who have a 
paramount interest in India.” 2 

But these conditions were rapidly passing. The 
French established a line ofsteamers between Marseilles 
and Alexandria in 1835, and under pressure from the 

1 Hoskins, British Routes to India, chapter v, passim. 

2 Auckland to Hobhouse, October 7, 1836 (Brit. Mus. Add. 
MSS, 36473 A, f. 91). 



Board of Control the East India Company ordered the 
building of two new steam vessels for the Bombay- 
Suez service. The development of the route, in accor- 
dance with the recommendation of the select com- 
mittee, was therefore assured. 

This, however, was not the only possible route. In the 
past Basra had been an active competitor of Suez, and 
steamers had proved so much more reliable on inland 
waters that men naturally wondered, in an age which 
had recently seen England covered with a network of 
canals, whether the best solution might not be found in 
joining the waters of the Orontes and the Euphrates, 
which, it was generally thought, would be a much 
simpler matter than cutting through the isthmus of 
Suez. In 1830-1 this route was being simultaneously 
surveyed by Chesney from Syria and by a party of the 
Company’s officers from India. The latter were seriously 
interrupted, and some of them were murdered, by the 
unruly Arabs who occupied the banks of the Euphrates. 
Chesney succeeded in the face of great difficulties in 
completing a preliminary survey. In 1 834 he was sent 
again with an expedition, carrying two flat-bottomed 
river steamers, which were to be put together on the 
upper waters of the Euphrates, to convey the expedi- 
tion down to the Persian Gulf. The Sultan’s farman was 
obtained, permitting the navigation of the Euphrates. 
After overcoming many obstacles, Chesney succeeded 
in assembling his two vessels on the Euphrates. One 
was sunk in a cyclone. The other actually reached 
Basra. But though the leader of the expedition re- 
mained enthusiastic about the possibilities of the route 
he had so laboriously surveyed, almost everyone else 
was convinced that, whatever its political importance 
and ultimate development, the Euphrates route could 
not compete with the Suez and Red Sea way to India. 1 

1 Cf. Hoskins, op. cit. chapter vn. 


However, the expedition had been inspired by a 
definite political object. Since the progress of Russia 
and the development of Muhammad ’Ali’s plans, the 
area had acquired a great importance; and a know- 
ledge of the facilities of transport which it offered was of 
the utmost political interest. It seems to have been 
opposed by the Russians as well as by Muhammad ’Ali. 
Ponsonby learnt at Constantinople that the Russian 
minister had conveyed a message to the Porte that the 
viceroy was ready to throw every possible obstacle in 
the way of the expedition, if that were the Sultan’s wish . 1 
Campbell at Alexandria believed that the Russian 
consul-general had endeavoured to prejudice the 
viceroy against the undertaking . 2 Great difficulties 
arose about work-people and supplies. Such opposition 
was but natural. The Russians did not want British 
influence established on the Euphrates, Muhammad 
’Ali feared we meant to build forts and occupy the 
river , 3 while from a revenue point of view he desired 
the Red Sea route to be developed rather than that of 
the Persian Gulf. Perhaps, too, he hoped that his 
opposition in a matter which the British seemed to 
have at heart would render them more amenable to his 
suggestions of independence. While therefore Ibrahim 
in Syria did all that could be done underhand to 
hinder Chesney’s progress, Muhammad ’Ali declined 
to send any orders to his son, until he should have 
received express orders from the Sultan . 4 Palmerston 
was very much annoyed. He wrote two stiff des- 
patches, in the second of which he observed that 
“His Majesty’s government are determined that the 
undertaking. . .shall not fail in consequence of the 

1 Ponsonby, November 6, 1835 (F.O. 78-256). 

2 Campbell, July 30, 1835 (F.O. 78-257). 

3 Sabry, op. cit. p. 299. 

4 Campbell, September 28, 1835 (F.O. 78-258). 



obstacles which bad faith in any quarter may oppose 
to it”. 1 

While Muhammad ’Ali was thus seeking to check 
British schemes for the survey of the ’Iraq rivers, the 
British Foreign Office was keeping a sharp lookout 
against any encroachments on the Sultan’s remaining 
territories. Muhammad ’Ali had, for instance, sought to 
add the district of Orfa to his Syrian territory, on the 
grounds that it had never been occupied by the Turks, 
that it was in a state of brigandage, that its inhabitants 
raided the country round Aleppo, that he would pay 
tribute for it, and that it had always formed part of the 
Aleppo pashaliq. 2 He was induced to withdraw from 
the district. In 1835 he occupied Deir on the Euphrates, 
doubtless with a view to control Chesney’s expedition 
more effectively. His excuse was that the nomad tribes of 
those parts raided his settled territories. 3 He was firmly 
warned not to encroach on the pashaliq of Baghdad. 

Whatever were the viceroy’s intentions about Bagh- 
dad, that city and Basra were regarded by the British as 
places of great moment. The occupation of Deir coin- 
cided with military activity in southern Arabia threat- 
ening to extend on to the Persian Gulf. “Great 
Britain”, Palmerston wrote, “would. . .think her in- 
terests directly concerned in preventing the authority 
of the Sultan from being shaken or interfered with at 
Baghdad”; and again, with regard to any movement 
towards Baghdad or the Gulf, “You will state frankly 
to the pasha that the British government could not see 
with indifference the execution of such intentions”. 4 

1 To Campbell, July [ ] and November 2, 1835 (F.O. 78-257 
and 258). 

2 Campbell, August 19 and October 6, 1834 (F.O. 78-246 and 
2 47 )- 

3 The same, December 21, 1835 (F.O. 78-258). 

4 To Campbell, Nos. 23 and 25, December 8, 1837 (F.O. 78- 


These words were no idle statement of mere views. 
Whatever might be the outcome of Chesney’s excursion 
down the Euphrates or the development of the Suez 
route, the Red Sea and the Persian Gulf formed direct 
avenues to India which Great Britain was resolved to 
guard with her own forces. 

The events which led to a further and sharper clash 
of Anglo-Egyptian interests began simply enough in a 
mutiny of Muhammad ’Ali’s troops in Arabia. The 
Syrian war had severely strained the viceroy’s finances. 
The pay of the troops in Arabia fell into arrears, and 
two Albanian officers exhibited great discontent. Early 
in 1832 the viceroy was writing to the governor of 
Hijaz that 5000 purses were being sent with which to 
appease the troops, but that “the two hogs of leaders” 
were to be persuaded to return to Egypt or else seized 
and sent thither in irons. 1 Neither measure bore fruit. 
The troops broke into open mutiny and their leaders 
defied the governor. One of their “ancient comrades” 
sent to restore discipline was forced to flee and returned 
to Cairo in disgrace, while the money sent to buy 
coffee on the pasha’s account was seized and plundered. 2 
At Jidda all public property and both the viceroy’s and 
private shipping were captured. 3 Late in 1832 the 
rebels were firmly established in the Yemen, 4 making 
Mokha their headquarters, where they seriously inter- 
rupted the Surat trade. 6 For the moment nothing 
effective could be done, but in the middle of 1833 
Muhammad ’Ali informed Campbell that he proposed 
to send an expedition to reduce Mokha 6 — a project 
with which the East India Company was in hearty 

1 To Hasan Aga, Ramzan 7, 1 247 (Abdine Archives) . 

2 Barker, July q 1 , 1832 (F.O. 78-214). 

3 The same, December 10, 1832 (ibid.). 

4 Campbell, April 16, 1833 (F.O. 78-227). 

6 The same, October 27, 1833 (F.O. 78-228). 

6 The same, June 11, 1833 (F.O. 78-227). 


sympathy. 1 At the close of the year an expedition set 
out, well provided with funds with which to buy off the 
rebels’ Arab allies. 2 These efforts were at last successful. 
The Arabs changed sides with their customary readi- 
ness, and the surviving rebel leader escaped on to a 
Company’s ship of war, while sixteen of his chief 
officers were beheaded. 3 

But although the Arab chiefs were willing to take 
Egyptian money for turning against the rebel troops, 
they were quite indisposed to allow Muhammad ’Ali to 
enjoy peaceful possession of country lying behind the 
southern Red Sea ports of Hodeida and Mokha. A 
long and troublesome war followed between Muham- 
mad ’Ali’s officers and the Assiri chiefs of Yemen, the 
former gaining small permanent advantage, while the 
state of war hindered trade. Even so late as 1838 
Campbell was still urging on the viceroy the unwisdom 
of seeking to reduce the Assiri tribes to submission 
instead of simply occupying the seaports and en- 
couraging the inland peoples to bring down their 
produce for sale. 4 

These operations in any case were of interest to Great 
Britain only in so far as they brought the Egyptian 
troops within striking distance of Aden; and indeed 
their lack of real success down to 1838 did not suggest 
any great probability that an effective control would be 
set up over the southern Red Sea littoral. But in the 
course of the year Muhammad ’Ali suddenly scored 
two considerable victories. On April 5 Ahmad Pasha 
contrived to slay 500 and capture 1000 of the hostile 
Assiri Arabs. 5 In the following month Kurshid Pasha, 

1 East India Company to Board of Control, August 9, 1833 
(F.O. 97-411). 

2 Campbell, December 5, 1833 (F.O. 78-228). 

3 The same, February 22, 1834 (F.O. 78-245). 

4 The same, March 20, 1838 (F.O. 78-342). 

6 The same, June 10, 1838 (F.O. 78—343). 


who had been despatched into the Wahabi country, 
reached Aneiza, about half-way across the peninsula in 
a direct line between Mecca and Basra. It was a place 
of some trade, frequented by merchants from Baghdad 
and Damascus, and likely to form a useful base for the 
further advance of the army. After some hesitation the 
local chief and principal people of the place came into 
Kurshid’s camp and made their submission. But an 
accident forced both sides to arms almost at once. In 
some personal dispute a Turk soldier shot an Arab 
inhabitant. In the riot that followed the Turk was cut 
to pieces, and some dozen killed on either side, while 
the troops were driven out and the gates shut against 
them. Kurshid was obliged to bombard the place for 
forty-eight hours before it surrendered. 1 This was 
followed up the next year by a further advance, which 
brought the Egyptian troops on to the shores of the 
Gulf itself. Early in 1 839 our agents in the Gulf re- 
ported the submission of A 1 Hassa and Katif, the 
territory lying along the western coast, and expected 
that the tribute “usually paid by the island of Bahrein” 
would be rigidly exacted by the governor whom 
Muhammad ’Ali had appointed over Najd. 2 Kurshid 
had written to our resident in the Gulf to announce his 
intention of occupying Bahrein, if necessary by force. 3 
The officer commanding the Egyptian troops who had 
advanced into A 1 Katif spoke (rashly indeed) to the 
British admiral who was visiting the Gulf of proceeding 
to reduce Basra and Baghdad, and Kurshid himself was 
reported only to be awaiting the arrival of more men 
from Medina to advance in force. 4 

These activities were at once most unwise and ex- 

1 Campbell, July 2, 1838 (F.O. 78-343). 

2 The same, April 6, 1839 (F.O. 78-373). 

3 Enclosures to Campbell, May 18, 1839 (F.O. 78-374). 

4 Maitland to Admiralty, April 7, 1839 (Ad. 1-219). 



ceedingly ill-timed. They involved a needless penetra- 
tion into an area where a strong British influence 
existed. The chief of Bahrein was one of “ the pacifi- 
cated Arabs of the Persian Gulf” (to borrow the extra- 
ordinary language of the Indian Political Department) 
who had signed the general treaty of 1820, and the 
government of India rightly decided to check this 
attack upon our position by ordering a firmer and 
more peremptory tone to be adopted towards Kurshid 
and his people, to be supported by the despatch of 
force and the cordial assistance of all chiefs who should 
resist the Egyptian demands. 1 2 Muhammad ’Ali’s ex- 
planation of his movements — that he only wanted to 
check the Wahabis, protect the Holy Cities, and pro- 
cure camels, 3 that he was being maligned by reports 
from Constantinople and Baghdad, 3 and that he did not 
desire any position on the Gulf 4 5 — had not even the 
merit of verisimilitude. 

And the movement was ill-timed, for it fell in with 
other unfortunate incidents, which, justly or unjustly, 
produced all the effect of design. In 1835 the Shah of 
Persia had contemplated the despatch of an envoy to 
Cairo. 6 In 1838 a member of the Persian mission at 
Constantinople visited the viceroy. 6 In the next year 
the Shah was reported to be going to send fifty young 
Persians to Cairo to be educated, 7 and early in 1840 a 
Persian messenger arrived with valuable gifts. 8 All 
these comings and goings may have been completely 
innocent. But they coincided with a time of Russian 

1 India to Bombay, secret, August i, 1839 (Ad. 1-220). 

2 Campbell, April 6, 1839 (F.O. 78-373). 

3 The same, July 11, 1839 (F.O. 78-374). 

4 Hodges, February 12, 1840 (F.O. 78-404). 

5 Ellis to Ponsonby, November 21, 1835 (F.O. 78-265). 

6 Campbell, March 19, 1838 (F.O. 78-342). 

7 The same, April 1, 1839 (F.O. 78-373). 

8 Hodges, February 6, 1840 (F.O. 78-404). 


predominance at the Persian court; when all the re- 
monstrances of the British agent had failed to dissuade 
the Shah from undertaking the siege of Herat, and 
when in 1838 an expedition had been despatched from 
Bombay to occupy the Persian island of Karak in the 

Other queer stories, too, were in the air. In 1835-6 
two men arrived at Alexandria from Constantinople, 
known respectively as Mahmud and Husain, whose 
antecedents and designs were not a little questionable. 
They were said to be mere adventurers, but had 
visited Russia in the character of emissaries from the 
court of Delhi. Mahmud was seen in Cairo and 
Alexandria, but then disappeared. Husain some 
months later landed with plague upon him. The 
English vice-consul, who redeemed a strong partiality 
for liquor by his knowledge of Turkish and Arabic, was 
asked to take charge of his baggage, as the sick man 
had 500 purses of money with him. Noticing the 
Indian accent with which Husain spoke, the vice- 
consul said he had met his companion when he passed 
through the country; Husain admitted that this was so, 
but was too ill to continue the conversation, and the 
next day fell into a delirium from which he never 
recovered. But his papers were secured from the 
lazaretto. They proved to be letters in Persian from the 
Grand Vazir to various Indian chiefs, together with a 
letter of introduction in Turkish from the Vazir to 
Muhammad ’Ali. 1 

The troubled political atmosphere inevitably in- 
vested Muhammad ’Ali’s advance on to the Gulf with 
a most suspicious air. The admiral on the East India 
station was therefore desired to visit the Gulf and to use 
all his influence to prevent any actual attack on 

1 Ponsonby to Campbell, March 31, 1836, and Campbell, 
January 18, 1837 (F.O. 78-319). 



Bahrein, though Auckland “would not in the absence 
of specific instructions from home authorize more 
decisive measures”. 1 The attitude at London was much 
more decided. Ponsonby at Constantinople was 
ordered to enquire whether Muhammad ’Ali’s con- 
quests had been made in accordance with the wishes of 
the Porte. 2 Campbell at Alexandria was ordered to tell 
the viceroy that instructions had been sent to Admiral 
Maitland to prevent the occupation of Bahrein, if 
necessary by force, 3 and, a few days before this despatch 
was written, Campbell, acting on earlier instructions 
and news received from India, had insisted on positive 
orders being sent to Kurshid to leave Bahrein alone. 4 

At the same period Muhammad ’Ali’s activity in 
Yemen had led to a very similar position at the entrance 
to the Red Sea. His victory over the Assiri Arabs in 
1838 gave him for the moment an undoubted ascen- 
dancy in the region known ironically as Arabia Felix. 
He was disposed to regard the ruler of Aden as a mere 
dependent, subordinate to the Imam of San’a, who had 
been compelled to recognise the viceroy’s authority. 5 
Alternatively he claimed it as formerly part of the 
Turkish empire. 6 

It was not possible however to treat such claims 
seriously. The imam had no doubt attempted from 
time to time to establish his power over Aden; but 
nothing like a real, permanent control had ever been 
exercised by him, and British recognition that he had 
any suzerain powers would have been a gratuitous 
absurdity. The Turkish claim was equally imaginary. 
Aden had once been occupied by the Turks in the days 

1 India to Bombay, secret, March 13, 1839 (Ad. 1-220). 

2 To Ponsonby, May 11, 1839 (F.O. 78-352). 

3 To Campbell, June 15, 1839 (F.O. 78-372). 

4 Campbell, June 15, 1839 (F.O. 78-374). 

6 Artin Bey to Boghoz Bey, March 22, 1838 (F.O. 78-342). 

6 Campbell, June 9, 1838 (F.O. 78-343). 




of Ottoman greatness in the sixteenth and seventeenth 
centuries. But, as it did not prosper under their rule, 
they abandoned it as valueless in 1630, and with their 
evacuation their rights disappeared. On various occa- 
sions in the recent past the British had come into 
friendly contact with the Sultan of Aden. When, in 
1 799, the British were anxious to block up the outlet of 
the Red Sea against the feared advance of Napoleon, 
they had occupied for a short time the island of Perim, 
that “stone in the sea belonging to the All -mighty, 
from which no revenue has or ever will be received”; 
but finding that hot and barren rock untenable, and 
having worn out or broken all their boring-rods in the 
fantastic hope of striking water, they removed for a 
while to Aden. There they were better off, and indeed 
it seemed almost a paradise in comparison with the 
little hell they had been occupying, while the sultan 
welcomed their presence and offered a constant supply 
of men for the Company’s military service. 1 In 1802 
Sir Home Popham actually concluded a treaty with the 
sultan. Valentia, who visited it in the course of his Red 
Sea travels, reported on it enthusiastically to Canning 
in 1808. After dwelling on the friendliness always 
shown by the sultan to the English, he continued, “It 
is the Gibraltar of the East, and at a trifling expense 
might be made impregnable”. 2 In 1822, when our 
resident at Mokha visited the place, he found it on the 
verge of falling into Muhammad ’Ali’s hands, the 
sultan having offered to admit a garrison and allow a 
small fort to be built on the eastern bay, provided he 
were to keep possession of the gates of the place, and 
exercise all civil and military authority within it. 3 Why 

1 From Murray, October 4, ap. Bombay Pol. Cons, December 

17 , 1799 - . „ 

2 Enc. in Valentia to Canning, September 13, 1808 (F.O. i-l). 

3 Hutchinson to Bombay, March 27, 1822 (I.O., Egypt and 
Red Sea, 7). 


Muhammad ’Ali did not take advantage of this oppor- 
tunity does not appear; and Salt, our consul-general, 
expected him to do so. 1 He certainly then missed his 
chance of obtaining complete control of the Red Sea, 
just as, a little later in the Greek war, he missed the 
most favourable chance he ever had of securing recog- 
nition of his independence. 

There matters remained until the project of the Suez 
route began to develop, and the need for coaling 
stations appeared. At first Socotra seems to have been 
chosen for that purpose, and the island was occupied by 
an expedition from Bombay in 1834-5. ® ut 11 proved 
most unsuitable. The high surf made landing difficult ; 
the island was full of malaria ; so the idea was aban- 
doned. 2 Already in 1829 Aden had been tried as a 
coal depot, on the occasion of the first experimental 
steam voyages from Bombay to Suez. But the Hugh 
Lindsay found she could not take in more th an 30 tons a 
day on account of lack of labour, a very extraordinary 
reason in the view of a modern traveller. 3 

Then very early in 1837 a Madras-owned vessel 
under English colours, the Darya Daulat, went ashore 
near by. She was carrying pilgrims and the large dona- 
tion that the Nawab of Arcot was in the pious custom of 
sending annually to Mecca. Such of the pilgrims as 
were not drowned were plundered by the Arabs, and 
all that could be got out of the wreck was seized under 
the direction of the sultan’s eldest son and sold in the 
bazaars principally by the sultan’s own agent. 4 On a 
detailed report of these occurrences, Sir Robert Grant, 
the Governor of Bombay, came to a prompt decision. 
“The establishment of a monthly communication by 

1 Salt to Hutchinson, December 7, 1822 (I.O., Egypt and Red 
Sea, 7). 

2 Low, Indian Navy, n 3 74. 

3 Idem, 11, 1 15. 

4 ParL Papers, 1839, xl, 42. 



steam with the Red Sea”, he wrote, “and the forma- 
tion of a flotilla of armed steamers renders it absolutely 
necessary that we should have a station of our own on 
the coast of Arabia, as we have in the Persian Gulf; and 
the insult which has been offered to the British flag by 
the Sultan of Aden has led me to enquiries which leave 
no doubt on my mind that we should take possession of 
the port of Aden .” 1 He was probably influenced also 
by the Egyptian expansion in Yemen. Auckland, how- 
ever, was less precipitate. He directed that repara- 
tion should be first required. If it were accorded, an 
amicable arrangement could be made about the coal 
depot; if it were refused, then further measures could be 
considered . 2 

Captain Haines of the Indian navy was therefore 
sent to discuss matters with the sultan. Early conver- 
sations went smoothly enough. After a long confabula- 
tion in which the sultan and his advisers “sat up all 
night with doors closed”, the sultan, who was in great 
fear that Lahej, the capital of his ancestral territory, 
might be seized by the troops of Muhammad ’Ali, 
decided to transfer the decaying port of Aden to the 
East India Company in return for an unspecified 
number of dollars. He even attached his seal to a 
document ceding Aden to the English. But at that 
point troubles arose. His eldest son was opposed to the 
change, and Haines did not venture to land to com- 
plete the settlement . 3 On this news Grant again urged 
the pressing need “to secure for the British Govern- 
ment, at the only moment when such a step is likely to 
be practicable for centuries, a possession which unfore- 
seen circumstances have placed within their reach ”. 4 
But the government of India thought the question 

1 Pari. Papers, 1839, xl, p. 54. 

2 Idem, p. 55. 3 Idem, pp. 56-61. 

4 Idem, p. 73. 


should be determined by the home authorities. 1 Action 
was therefore delayed until the arrival in August of de- 
spatches from the secret committee, 2 on which Auckland 
authorised the Bombay government to proceed. 3 Haines 
w^s at once sent off, with a draft treaty in his pocket, and 
an escort of thirty men of the Bombay Europeans, for fear 
lest Muhammad ’Ali should swoop down upon Aden 
while a more imposing force was under preparation. 4 5 

Haines reached Aden on October 24, and, as anyone 
with a knowledge of the East might have anticipated, 
the inadequacy of his force encouraged the sultan’s son 
earnestly to press his father not to accede to the English 
proposals. In this he succeeded. Plundered goods from 
the Darya Daulat, which had been given up and stored, 
were not allowed to be removed. A little later the 
Arabs, waxing bold, fired on the English boats. Haines 
retired to a small island to await reinforcements. They 
arrived on January 16, and two days later the town was 
carried by assault. 

Sir Charles Malcolm, who as superintendent of the 
Bombay Marine had originally suggested “getting the 
cession from the Sultan. . .instead of getting permission 
for a coal-depot held at the will of an avaricious and 
capricious Arab chief”, was delighted with the issue of 
events. “The harbour of Aden”, he wrote, “with its 
capacious bay only open to the southward, surpasses 
my expectations. We could not have desired anything 
better to answer all our purposes.. . .” 6 

1 Pari . Papers , 1839, xl, p. 76. 

2 Secret Committee to India, May 30, 1838 (I.O., Board’s 
Drafts No. 9). 

3 India to Bombay, September 3, Bombay Sec. Cons., October 
3, 1838, No. 650. 

4 Governor’s Minute, Bombay Sec. Cons., September 5, 1838, 
No. 488. 

5 Malcolm to Locke, January 18, 1839 (Brit. Mus. Add. MSS, 

36470, f. 57). 


This British occupation of Aden was a sharp disap- 
pointment to Muhammad ’Ali — far sharper, probably, 
than our insistence on his withdrawing from the 
Persian Gulf. It traversed his commercial as well as his 
political views. It was expected, for instance, although 
the expectation was never realised, that the whole trade 
of Mokha coffee would be removed from Mokha to 
Aden , 1 which would have meant the loss of a very 
valuable monopoly. The Egyptian commandant com- 
plained that the Mokha customs had already vanished . 2 
Foreign nations, especially the French and the Russians, 
disliked a change which could not but strengthen the 
British position in the East. “ I feel certain”, Campbell 
wrote, “that they have worked, and will continue to 
work on the mind of Mehemet Ali, in giving him false 
and erroneous impressions of our views in the possession 
of Aden .” 3 But with whatever inward feelings the 
viceroy regarded this approach of British authority, he 
carefully avoided protest and limited himself to a state- 
ment of wishes and hopes. On learning that the Indian 
governments had decided to suspend action till they 
could receive orders from home, Muhammad ’Ali 
observed “ that he hoped the Indian government would 
be persuaded that Aden formed part of the Yemen . . . , 
and that he trusted the Indian government could not 
for a moment doubt with what pleasure he would 
permit the establishment of a coal depot there, as well 
as in any other part of his governments ”. 4 His nearest 
approach to a formal protest was the suggestion 
that it was not reasonable for us to have approved his 

1 Campbell, November i, 1837 (F.O. 78-321). 

2 General commanding Yemen to Muhammad ’Ali, February 
12, 1838 (F.O. 78-342). 

3 Campbell, March 27, 1838, and April 8, 1839 (F.O. 78-342 
and 373). 

4 The same, June 9, 1838 (F.O. 78-343). 


expedition into Yemen and then to seize one of its 
dependencies . 1 

But Palmerston was not to be appeased. With angry 
pencil he heavily underscored Muhammad ’Ali’s 
reference to his governments, as though the possessive 
adjective were in itself treason against the viceroy’s 
august master, Great Britain’s trusted ally. As to our 
having sanctioned the expedition into Yemen, he tartly 
replied that we had raised no objection to Muhammad 
’Ali’s reducing his mutinous troops, but the expedition 
had been despatched long before our assent had been 
received . 2 When the viceroy suggested that it might be 
well to avoid frontier incidents by recalling his troops 
from Yemen, the Foreign Secretary retorted that he had 
no desire that the Egyptian occupation of Yemen 
should continue, “but on the contrary would be better 
pleased by any overt act which should show that the 
pasha is engaged in improving the administration of 
the provinces confided to his government instead of 
employing the energies of his mind and the resources of 
the country he governs in aggressive expeditions against 
neighbouring districts ”. 3 And even before Aden had 
been actually occupied, he warned the viceroy that “any 
hostile attempt . . . against Aden will be an attack upon a 
British possession and will be dealt with accordingly ”. 4 

From this time onward, as relations became tenser in 
consequence of the developments in Syria, Aden con- 
tinued to contribute occasions of bitterness. Muham- 
mad ’Ali was warned not to interfere with the Arab 
chiefs in the neighbourhood of the new British posses- 
sion . 5 A little later it was suggested that the viceroy 

1 Campbell, April 17, 1838 and enc. (F.O. 78-342). 

2 To the same. May 12, 1838 ( ibid !.). 

3 To the same, May 24, 1838 (ibid.). 

4 To the same, June 8, 1838 (F.O. 78-343). 

6 To the same, May 11, 1839 (F.O. 78—372). 


would do well to act on his own suggestion and evacuate 
Yemen altogether . 1 Then came reports that the 
Egyptians were stirring up chiefs to attack Aden, and 
the tactless, stupid consul-general who had replaced 
Campbell swallowed the whole story as literal and 
exact fact , 2 and, even when Muhammad ’Ali had 
wholly withdrawn from the country, reproached him 
with the conduct of those who had succeeded to his 
authority . 3 

Altogether the period which we have been consider- 
ing, falling between the first and second Syrian wars, 
illustrates alike the strength and weakness of Muham- 
mad ’Ali’s policy and the actual inconsistency of its 
several parts. He saw quite clearly the great impor- 
tance to him of British friendship. He always seems to 
have desired it. He was constantly seeking the means 
by which he could increase the value of his co-operation 
in English eyes. But there he quite misjudged his own 
position and the position of Great Britain. English and 
Egyptian interests were indeed closely interwoven. The 
development of the Suez route to India, gave us the 
strongest possible interest in preserving Egypt from any 
foreign domination (except perhaps our own), in pro- 
moting the stability of her government and the pros- 
perity of her people, while British sea power rendered 
Britain the best possible ally for a country that could 
hardly be attacked but from the sea. An Anglo- 
Egyptian alliance was therefore a thoroughly sound 
conception. But from our own point of view there was 
a wide difference between Muhammad ’Ali, Pasha of 
Egypt, promoting order, justice, and education in the 
valley of the Nile, and Muhammad ’Ali, using up the 
population of his country in order to conquer Arabia 

1 To Campbell, September 13, 1839 (F.O. 78-372). 

2 Haines to Hodges, February 10, 1840 (F.O. 78-404). 

8 Hodges, February 22 and July 6, 1840 (F.O. 78-404 and 405). 


and Syria, to spread his power eastwards to Basra and 
southwards to Aden, and threatening to convulse 
Europe by overturning the Turkish empire. Nothing 
could persuade Palmerston — and indeed in this he was 
entirely right — that British interests needed the support 
of a great military power in the Near East, such as 
Muhammad ’Ali, and especially his son Ibrahim, had it 
in mind to build up. British interests were to be pro- 
tected by British arms. So the viceroy’s expansion 
necessarily led to a conflict of interests and an opposi- 
tion of policies. But equally clearly there was not, what 
some recent Egyptian writers have suggested, the least 
hostility to Egyptian greatness. Muhammad ’Ali could 
be as great as he pleased within the natural geographi- 
cal limits of Egypt. But he was not to endanger the 
peace of Europe or undertake duties for Great Britain 
which Great Britain felt very well able to discharge for 
herself. Palmerston wisely and rightly preferred to 
establish his country’s power in the Persian Gulf and at 
the entrance of the Red Sea rather than allow any 
other, however friendly his protestations, to occupy 
regions that were clearly destined to a high importance. 

Chapter VI 


The tendencies generated by the questions of the Red 
Sea and the Persian Gulf unhappily corresponded in 
direction and effect with tendencies generated by de- 
velopments in the Levant. The settlement reached by 
the Peace of Kutahia 1 had been no true settlement at 
all, for it had left every party dissatisfied and desirous 
of further changes, and in particular was this true of 
the two principals in the struggle. At Constantinople 
the Sultan, Mahmud, and his sar-’askar, Khusrau, 
were bent, the one on recovering Syria, the other on 
humiliating his old rival. On the other side Muham- 
mad ’Ali had indeed secured a great accession of 
territory, but he held it on the least satisfactory of 
terms — on grants annually renewable and so annually 
revocable by the Sultan. The pasha was growing old. 
He could not look for many more years of life ; and 
what would become of his pashaliqs and of his family 
when he died? His disappearance from the scene 
would be the certain signal for a strong endeavour to 
bring not only Syria but Egypt as well back under the 
Sultan’s immediate control. His family would be per- 
secuted in revenge for his conduct. The countries in 
which he had laboured to improve administration and 
to spread knowledge would be given over to pashas of 
the old school whose only object would be to shear the 
people as closely as possible before they were recalled. 
As far as Muhammad ’Ali could judge, neither his 
family nor his reforms were likely long to survive their 

1 Vide pp. 1 2 1-2, supra. 


founder, his own name would speedily pass away and 
be forgotten, and the work to which he had devoted his 
life and energy would vanish as though it had never 
been. The older he grew, the more deeply was it borne 
in upon him that his work was still completely un- 
secured against future events. 

The relations of the Sultan and the pasha imme- 
diately after the war showed how hollow had been their 
reconciliation. There was the question of the tribute. 
Even when the amount payable had been discussed and 
settled, the Sultan still claimed arrears which Muham- 
mad ’Ali absolutely refused to pay. While this matter 
was pending, advantage was taken of the marriage of a 
princess of the Sultan’s family to send a special envoy 
to the capital, ostensibly on a mission of congratula- 
tion, but in fact with a much more serious object. The 
envoy was accompanied by a suite of twelve persons : 
he was ordered to display all “the magnificence of a 
vizir”; he was to distribute a million piastres in 
presents . 1 But he was also to represent to Mahmud 
that so long as Khusrau continued in the Divan, he 
would always be seeking to misrepresent the pasha’s 
conduct, and that “if the sultan would only consent to 
remove the seraskier 2 from his councils, he (the pasha) 
would not only pay the tribute regularly . . . , but he 
would also pay a great part of what the Sultan de- 
manded as arrears”. It was expected that various 
influential enemies of Khusrau would be gathered to- 
gether at Constantinople for the marriage, so that the 
opportunity appeared promising . 3 However, the mis- 
sion not only failed completely in its object but became 
the occasion of slights directed at its sender. The envoy, 
Habib Effendi, was not allowed to wear a flag on his 

1 Campbell, April 27, 1834 (F.O. 78-245). 

2 I.e. the sar-’askar. 

3 Campbell, May 10, 1834 (F.O. 78-245). 


boat; he was not allowed an awning to shelter him 
from the sun ; his rowers were not suffered to pull in the 
manner commonly used when conveying a person of 
distinction ; the chief people at the capital were afraid 
to pay him any public visits, and would only receive 
him in private; and the Sultan himself showed dis- 
pleasure when the crew of the frigate that had con- 
veyed Habib to Constantinople manned the yards and 
cheered, European fashion, in acknowledgment of a 
gift of 50,000 piastres. 1 

In the course of the year 1834 the question of the 
tribute was at last arranged, on the basis of the 
customary rates without arrears. But that hardly signi- 
fied any real improvement in the difficult and anoma- 
lous relations between Mahmud and Muhammad ’Ali. 
The former, for instance, took every occasion to stir up 
trouble in Syria. Ibrahim had introduced conscription 
into the province along with measures of protection for 
the Christian inhabitants. Both had excited great dis- 
content; a revolt broke out in the country around 
Jerusalem; and matters became serious enough for the 
viceroy to visit Syria in person. There was no doubt 
that the revolt had been encouraged by agents from 
Constantinople, whose preaching might be reasonably 
inferred from an incident that occurred at Nablus. 
There a Turk mounted the minar of the mosque, and 
cried aloud: “Does the Muslim religion exist no 
longer? Is it dead? Are we not Turks? Let every man 
who loves the Prophet take up arms against that man 
without faith, the Giaour, Ibrahim Pasha ! That 
drunkard who always drinks spirits and wine, who eats 
pork and every dirt that comes from the sea [alluding 
to Ibrahim Pasha’s eating turtle and other sea-fish for- 
bidden by the Mahometan religion], the same as the 
Christians do, who lives in the convents with the 
1 Campbell’s Journal, July 15 and 16, 1834 (F.O. 78-246). 


priests and prays with them, but never goes to the 
mosque”. 1 However, vigorous measures were used. 
Three rebel chiefs brought in to Muhammad ’Ali were 
beheaded out of hand. 2 The rebellious districts were 
disarmed. The conscription was enforced. The pasha’s 
power remained unshaken. 

But the general position was full of danger. Each 
party heartily distrusted the other. Each began to 
prepare for the coming and decisive conflict. “Every- 
thing in Syria”, wrote the British consul at Aleppo, “is 
military and every measure is being taken to increase 
the strength of the army.” The defiles of the Taurus 
were fortified. The pasha’s troops were concentrated 
behind his northern frontier. The position was the same 
on the other side of the border. Some 9000 men were 
assembled at Koniah. 3 

What aroused special attention and indeed particular 
disapproval in England was the conscription by which 
the pasha’s military forces were kept up to strength and 
enlarged. This was an unwelcome novelty in Syria. 
The old pashas had never dreamt of such a thing. They 
had generally maintained Albanian or other foreign 
mercenaries, despising the military qualities of the 
native Syrians. 4 Muhammad ’Ali resolved to use them. 
He had no census to rely on, nor was any census 
possible. Nothing but Roman discipline could enforce 
a measure universally considered as unlucky in itself 
and certain to lead to new taxation. On these points 
the Syrians still cherished the beliefs they had held in 
the days of Augustus. The only method by which con- 
scription could be enforced was to order a levy of a 

1 Campbell’s Journal, June 30, 1834 (F.O. 78-245). The 
translation has been modified. 

2 The same, July 17, 1834 (F.O. 78-246). 

3 Werry to Campbell, June 2, 1835 (F.O. 78-257). 

* Cf. Memoirs of Lady Hester Stanhope, n, 1 13 sqq. 


specified number of men in a given area, and seize 
them by force. The Syrians, who much preferred 
suffering the insults of undisciplined mercenaries to 
serving in arms themselves, did everything in their 
power to avoid being caught. In Aleppo, for instance, 
when a levy of 10,000 men was ordered in 1833, men of 
military age went into hiding. Some took refuge within 
consular precincts, and their parents were bastinadoed 
within sight of the windows in the hope of making the 
refugees come out. Finally the heads of the quarters of 
the city were compelled to make up the quota. 1 In 1835 
much the same was repeated with the same passive 
resistance. At Beyrout the mosques were surrounded 
and the males seized. At Aleppo houses and shops 
were closed. Trade ceased. For two days neither bread 
nor meat nor any other kind of provision could be 
procured. Many fled to the villages at the foot of the 
Taurus. Others, disguised as women, made their way 
over the frontier into the Sultan’s territories, only to 
find that Mahmud was following Muhammad ’Ali’s 
example and himself carrying out a conscription with 
great severity. 2 

These events were reported with much exaggeration, 
and excited much unfavourable comment. Campbell 
was directed to suggest confidentially but unofficially 
to Muhammad ’Ali that if conscription was really 
needed it should be made on regular lists, on an or- 
ganised system, instead of men being seized promis- 
cuously by military force “much in the manner in 
which a given number of wild animals would be caught 
out of a herd in the desert”. 3 These general humani- 
tarian sentiments were reinforced in certain directions 
by more particular interests. Sometimes this was justi- 

1 Campbell, February 3, 1834 and end. (F.O. 78-245). 

2 The same, February 18, 1836 (F.O. 78-282). 

3 To the same, December 8, 1837 (F.O. 78-318). 


fied, as when in 1835 the troops at Beyrout seized 
persons in the employment of the consulates; and on 
this occasion Muhammad ’Ali sent Colonel Seve 
(Sulaiman Pasha) to make a special enquiry, and 
invited the consuls-general at Alexandria to depute a 
person to accompany him. 1 Sometimes, however, 
Palmerston blazed up over premature and inaccurate 
reports. In 1835 he heard that Christians had been 
included as conscripts. “Europe”, he wrote, “has a 
right to hope that the Christian subjects of the Porte 
who inhabit the countries which the Sultan has placed 
for the present under the administration of Mehemet 
’Ali shall be exempted from that new conscription with 
which the pasha thinks proper to harass and afflict the 
Mahometan population whose interests and welfare 
have been committed to his keeping.” 2 But Campbell, 
ignoring his indignant irony, was able to reply that no 
Christians had been conscripted. He had recently 
made a tour in Syria, where he had seen pilgrims with 
a cross tattooed on their arms, and he had learnt from 
them it was a common custom, which not only distin- 
guished them from Muslims but also freed them from 
the fear of compulsory enrolment. 3 

But while humanitarian ideas and Christian sym- 
pathy contributed to the disapproval with which the 
powers, and particularly Great Britain, regarded the 
Syrian conscription, political considerations made it 
positively alarming. War between the Sultan and the 
pasha threatened the reappearance of the Russians and 
the confirmation of Russian influence at Constantinople 
under the Treaty of Unkiar Skelessi; and then Great 
Britain would have either to acquiesce in Russian 
predominance in the Straits or take up arms to destroy 

1 Campbell, September 5, 1835 (F.O. 78-258). 

2 To the same, May 9, 1836 (F.O. 78-281). 

3 Campbell, July 10, 1836 (F.O. 78-282). 


it. Neither alternative could be viewed with compo- 
sure. Muhammad ’Ali must therefore be restrained 
from attacking the Porte, or, if war could not be pre- 
vented, Great Britain must join Russia in supporting the 
Sultan. Repeated remonstrances were therefore ad- 
dressed to the viceroy. At the end of 1837 Campbell 
urged upon him that the powers would not permit him 
“to keep up large armaments which could have no 
other effect than to embroil him with the Sultan and 
prevent the pacification of the east ’’A Palmerston sent 
loud warnings “of the evil consequences which will 
infallibly result to himself if he recommences an attack 
upon any part of the sultan’s dominions. You will also 
represent to the pasha that his extensive conscription, 
his active military preparations, and his concentration 
of troops in Syria, are all calculated to excite great 
distrust as to his intentions with respect to the Porte”. 2 
But to all remonstrances Muhammad ’Ali had an 
answer that was difficult to rebut. Mahmud was re- 
organising his own army; German officers, including 
the famous von Moltke, were being employed to train 
and discipline them; fifty-one new regiments, it was 
reported, were being formed. But the Porte was in- 
volved in no foreign war and had no internal rebellion 
to crush. The preparations must then be directed 
against Egypt, and the viceroy’s measures were dictated 
by the heartiest desire for peace 3 — a mere eastern 
version of Si vis pacem . . . . 

Both Great Britain and France found this rejoinder 
very irritating. Their consuls-general were instructed 
to make the strongest representations, and twice in 
March, 1838, Palmerston wrote on the subject, first 
demanding a categorical explanation of Muhammad 

1 Campbell, December 12, 1837 (F.O. 78-321). 

2 To the same, February 6, 1838 (F.O. 78-342). 

3 Campbell, February 7, 1838 ( ibid .). 


’Ali’s intentions, 1 and later warning him of the 
dangerous consequences of war. “You will point out 
to the pasha”, he continued in this second despatch, 
“that he ought to be sensible that his talents and 
energies, great as all the world know them to be, will 
find ample scope for their exertion in establishing a 
good system of administration in the countries already 
subject to his rule.” 2 

But in spite of his fine words and moral indignation 
Palmerston was not taking, perhaps from his position 
he could not take, a fair view of the situation. He was 
demanding of the viceroy what could not possibly be 
conceded save to force. Campbell had suggested a 
much juster attitude when at the end of 1837 he had 
written, “I cannot but feel that if it were possible for 
Mehemet ’Ali to be secured against any aggression on 
the part of the Porte, and that he were at the same 
time obliged to reduce his army and navy to a fixed 
standard, and prohibited from raising conscriptions in 
any part whatsoever of his governments, the beneficial 
change of these measures would be speedily visible in 
every part of the country”. 3 This seems entirely true. 
Nothing but some such guarantee could free the viceroy 
from the need of arming unless he was prepared to 
hand over to the Sultan any of his governments that the 
Sultan chose to demand. Unfortunately the Russian 
position made any such guarantee impossible. Palmer- 
ston therefore adopted the official theory that Muham- 
mad ’Ali was merely a servant and minister of the 
Sultan, that the Sultan had every right to require the 
restoration of his territories at any time, and that the 
viceroy’s warlike preparations were illegal, disloyal, 
treasonable. So they were in the theory of the Otto- 

1 To Campbell, March 16, 1838 (F.O, 78-342). 

2 To the same, March 29, 1838 (ibid.). 

3 Campbell, December 21, 1837 (F.O. 78-321). 




man empire. But where lies the point at which worn- 
out fictions such as this cease to bind? In India the 
government of the East India Company had resolved 
that it was freed from all obligation to the Moghul 
Emperor as soon as he quitted their protection and 
cast in his lot with their possible enemies, the Marathas. 
Every reasonable person had always held that they 
were justified. Muhammad ’Ali’s position was not 
wholly unlike that of the Honourable East India 
Company. The main difference lay rather in political 
environment than in political principle. Warren 
Hastings’ rejection of the authority of Shah Alam did 
not imperil the peace of Europe; Muhammad ’Ali’s 
rejection of the authority of Sultan Mahmud would 
endanger it. While therefore the great pasha deserved 
sympathy in his endeavours to make permanent the 
reforms which he had introduced and to save them 
from the blight of Turkish administration, there still 
remained, quite apart from the unconvincing argu- 
ments on which Palmerston’s official case was rested, 
strong and valid reasons for his. policy. In this world 
major interests must be suffered to prevail, and the 
consolidation of Muhammad ’Ali’s power or even the 
maintenance of his reforms could not be reckoned to 
outbalance the evils of a general war. “The great 
object of the British Government”, Palmerston had 
said in 1833, and it still was true, “is the maintenance 
of peace . . . ; we are averse to any great changes in the 
relative distribution of political power because such 
changes must either be brought about by war or must 
have a tendency when effected to produce war.” 1 

The nature, efficacy and ideas of Muhammad ’Ali’s 
administration must be reserved for treatment in a 
separate chapter. But, whatever their real value, it was 
an important element in Palmerston’s policy that he 
1 To Campbell, October 2, 1833 (F.O. 78-226). 


regarded them with a large measure of distrust. The 
liberal statesman thought the benevolent, humani- 
tarian, enlightened objects which men said the pasha 
had set before himself completely inconsistent with the 
violent seizure of men to serve in the pasha’s armies ; 
the whig landlord could not conceive just govern- 
ment compatible with expropriation; while western 
economists condemned with one voice the trade mono- 
polies which the pasha had set up. These general 
reasons strongly indisposed him to sympathise with 
Muhammad ’Ali’s claims and views. When Campbell 
alludes to the viceroy’s protection of property, Palmer- 
ston comments, “ Except that of the people whom he 
governs”, and, when the consul-general refers to the 
pasha’s benevolence, the foreign secretary adds, “I.e. 
war and conquest and plunder and conscription and 
monopoly ”. 1 These expressions, valueless as an esti- 
mate of Muhammad ’Ali’s work, undoubtedly help to 
explain Palmerston’s conduct in the crisis that was 

Nor were these views merely the product of general 
ideas and the irritation created by the pasha’s incon- 
venient activity. The administration of Syria, as I shall 
show later, had been less successful than the administra- 
tion of Egypt, and its defects had been artfully ela- 
borated by sinister interests. The decay of Turkish 
administration, the neglect of the pashas, the growing 
inferiority of Turkish military power, and the con- 
sequent timidity of the Divan in matters of external 
policy, had encouraged the growth of extraordinary 
abuses of the Turkish capitulations. The consuls 
claimed to be exempt from all but certain fixed dues 
and to be able to extend this privilege to all whom they 
employed and to all whom they recognised as their 
nationals. Layard records that at Salonika most of the 
1 Campbell’s Report on Egypt (F.O. 78-408 B). 



consuls lived on the sale of passports and protections to 
the native Christians . 1 In Syria these abuses had been 
carried on without check. “The consuls and agents”, 
Campbell wrote, “ used to protect an unlimited number 
of rayas of the country, under the denomination of 
brokers to the merchants, honorary dragomans, etc., 
and these protections were sold by them to the rayas, 
some of whom being rich were ready to give large sums 
for a protection which withdrew them from Turkish 
jurisdiction .” 2 Lady Hester Stanhope, with no better 
excuse than her own autocratic temperament, gave no 
less than seventy-seven protections, some to persons of 
wealth, and nearly all to persons who were not in her 
service or at all events drew no pay from her. The consuls 
too were always issuing certificates declaring goods in 
the Turkish customs houses to be for their personal use 
(and so exempt from duty or examination) when every- 
one knew that they were merely covering goods belong- 
ing to native merchants . 3 The establishment of Mu- 
hammad ’ Ali’s rule in Syria, involving the introduction 
of conscription, greatly enhanced the money value of 
consular protection. Colonel Seve, who had been sent 
to enquire into the violation of consular buildings , 4 
brought back a severe report, confirmed by the consuls- 
general’s agent who had accompanied him. The drago- 
mans were mostly rich native merchants, who, far 
from being able to interpret for the consuls, knew no 
language but Arabic; the janissaries kept shops or 
followed a trade ; the clerks were merchants, and some 
rich. Most of these nominal employees were either 
above discharging their duties or incapable of doing so, 

1 Layard, Autobiography, n, 25. The reader will notice the 
similarity to the practice of the East India Company’s servants in 
Bengal between 1757 and 1765. 

2 Campbell, June 19, 1834 (F.O. 78-245). 

3 Ibid. 

4 Vide p.159, supra. 


but they paid handsomely for their offices, and the more 
so because consular protection was held to extend not 
only to their employees, but to their employees’ 
families and servants. 1 Campbell himself furnished 
specific examples of abuses occurring within his own 
knowledge. At Beyrout in 1836 he had found the 
British consul protecting a cargo of wheat which proved 
to have been shipped by one Greek and consigned to 
another. 2 When therefore on Muhammad ’Ali’s legiti- 
mate complaints the consuls-general at Alexandria 
advised that these profitable protections should be 
limited 3 at the very moment when they promised to be 
more profitable than ever, the irony of events bit deep 
into the consuls’ hearts. Far better, they felt, had been 
the easy days of Turkish misgovernment, when (for a 
consideration) a raya could easily obtain Russian or 
French or British status, than the regular system of 
profitless reforms that were being introduced from 
Egypt. Their reports inevitably reflected their lacerated 
feelings and empty hands. Campbell had frequent 
reason to remark “the extreme eagerness and avidity” 
with which some of them seized upon every trifle 
likely to indispose His Majesty’s Government against 
the viceroy. They reported monopolies which had never 
been established. 4 They demanded that British drago- 
mans, who were Levantines of low birth and small 
education, should be received with the same distinc- 
tion as French dragomans, who were educated Euro- 
peans, bearing their sovereign’s commission, and en- 
titled in due course to promotion to consular rank. 5 

1 Campbell, November 22, 1835 (F.O. 78-258). 

2 The same, July 19, 1837 (F.O. 78-320). 

3 The regulations are given in Campbell, November 22, 1835 
(ut supra). They were ratified promptly by Austria, France and 
Russia. The same, April 4, 1839 (F.O. 78-376). 

4 Campbell, December 3, 1836 (F.O. 78-284). 

6 The same, September 22, 1837 (F.O. 78-320). 


1 66 

And one even got up a bogus petition against the 
Egyptian authorities, in support of indefensible abuses. 1 

Nor did they lack a convenient channel of communi- 
cation with the British embassy at Constantinople. The 
second dragoman there, Richard Wood, was brother- 
in-law of Consul Moore at Beyrout. His temper may 
be judged from the following incident. When Campbell 
was touring in Syria in 1836, he met this man at 
Beyrout, and listened to a harrowing account of 
Ibrahim’s cruelties in suppressing a recent revolt, and 
in particular of his having burnt to the ground no less 
than thirty villages. What were their names? asked 
Campbell. Wood did not know. Had he seen them? 
No, but he had been told so. Campbell very properly 
begged Consul Moore to verify the statement. But 
although Moore was unable to do so, Wood neverthe- 
less reported the story as a fact to Ponsonby, the 
ambassador. 2 These consular reports must have fallen 
in aptly with Palmerston’s frame of mind, ruffled by the 
clashes of policy described in the last chapter, and 
thoroughly irritated by the threat to the peace of 
Europe implicit in the relations between Muhammad 
’Ali and the Sultan. He was thus strongly predisposed 
by his view of the European situation and his disbelief 
in the reality of Muhammad ’Ali’s reforms to support 
the Sultan rather than the pasha. 

The attitude of France was somewhat different. The 
French had never regarded the Ottoman empire as 
sacrosanct. They had not hesitated to detach Algiers 
from it. They had on one occasion sent a minister 
plenipotentiary direct to Alexandria. 3 Louis-Philippe 
had in private conversation spoken of Muhammad 
’Ali’s independence as a thing certain to be achieved in 



^ampoeii, ^ctooer 9, io; 
The same, July 31, 1836 


time. 1 The French had provided officers for the army, 
the marine, the dockyard and arsenals, of the pasha. 
French financiers had offered him a loan. 2 The French 
consul-general maintained close and friendly relations 
with him. Like the English, the French wanted peace 
in Europe; but, unlike the English, they were disposed 
to seek it by preventing the Sultan from attacking 
Syria rather than by preventing Muhammad ’Ali from 
strengthening himself against the Turk. Their first plan 
was to reconcile the two. “To France belongs the duty 
of uniting the two halves of the empire”, wrote 
Mimaut, the French consul-general. In the course of 
1836 the French ambassador at the Porte was directed 
to offer French mediation. Apparently the French 
were willing to guarantee Muhammad ’Ali’s position 
for life provided he reduced his army and marine by 
half, which would allow the Porte to do the like ; 3 and 
the ambassador, on the eve of returning to Paris to 
confer with Thiers and the French consul-general with 
Muhammad ’Ali, declared to the Reis Effendi that the 
Porte’s hostility against the viceroy must be abandoned. 4 
As a result of these suggestions and of obscure discus- 
sions between the French authorities, the Porte, and 
agents of Muhammad ’Ali, 5 it was decided to send 
Sarim Effendi on a special mission to Egypt. But this 
was only one more example of the Porte’s habitual bad 
faith. At the moment when it appeared to be comply- 
ing with French wishes, it was writing (with perhaps 
equal sincerity) to its minister in London that it was 
seeking to content the French ambassador “without 

1 Campbell, May 30, 1833 (F.O. 78-227). 

2 The same, October 12 and 24, 1833 (F.O. 78-228). Cf. 
Sabry, op. cit. pp. 31 1 sqq. 

3 The same, October 30, 1836 (F.O. 78-284). Cf. Sabry, op. cit. 
P- 3 * 9 - 

4 Sabry, op. cit. p. 320. 

5 Campbell, December 20, 1836 (F.O. 78-284). 


opening the door of our mind ” and that Great Britain 
was the only power on whom the Turks could rely . 1 
When Sarim reached Alexandria, Muhammad ’Ali 
found that he was sent merely to flatter him and sound 
his views. A couple of days after his emergence from 
the quarantine which fear of plague inflicted on all 
arrivals from Constantinople, Muhammad ’Ali re- 
ceived Campbell in audience. Their talk chanced to 
turn on the fits of madness that gossip attributed to the 
Emperor Nicholas. “ I do not believe ”, said the pasha, 
“ that he is the only sovereign who is so. Mine also does 
not appear to be right in his head,” for he had sent 
an envoy to arrange for the co-operation of Cairo 
and Constantinople without empowering him to offer 
terms . 2 In later discussions Sarim proposed an heredi- 
tary tenure of Egypt and Acre, while the pasha in- 
sisted that the offer must include all his existing 
territories . 3 The mission was therefore as complete a 
failure as the Divan had intended. But the obstacles to 
a good understanding had been rendered more evident 
than ever . 4 The bad faith of Constantinople was 

The French share in this abortive overture was well 
known ; and it was generally thought that but for their 
encouragement and support Muhammad ’Ali would 
have paid more attention to Palmerston’s representa- 
tions. They certainly were eager to keep the pasha on 
his guard against what they believed to be a consistent 
British hostility . 5 And it seems to have been Metter- 
nich’s policy to promote British suspicion of French 
plans. All that his agents could discover, steal or invent 

1 Sabry, op. cit. pp. 320-1. 

2 Campbell, January 21, 1837 (F.O. 78-319). 

3 Idem, and January 27, 1837 (ibid.). 

4 The same, April 11, 1837 (ibid.). 

6 E.g. instructions to Cochelet as consul-general, September 1 2, 
1837, ap. Sabry, op. cit. pp. 325 sqq. 


in this connection was freely placed at the disposal of 
their British colleagues. De Laurin, the Austrian 
consul-general, communicated to Campbell not only 
his despatches to his own furtive Foreign Office, but 
also documents which he had “conveyed” from the 
French consulate. He once showed to Campbell, for 
instance, a letter from Colonel Seve to Mimaut, with 
marginal comments and signature in Mimaut’s own 
hand. 1 Nor were the French obstructed in the execu- 
tion of their policy only by dishonest Turks or foreign 
mischief-makers. With that lack of disciplined sub- 
ordination which their agents in the East had ever 
displayed, Roussin at Constantinople or Sebastiani at 
London might hold language quite inconsistent with 
the views of the French cabinet. 2 

At last in 1838 Muhammad ’Ali, finding that he had 
in no wise benefited by the well-meant French en- 
deavours, resolved upon bringing matters to a crisis. 
This was fancied at the British embassy at Constanti- 
nople to be the result of Russian advice. The idea had 
obsessed the embassy for some years, and there are 
grounds for thinking that it was promoted by vexed 
Levantine consuls. 3 Campbell did not believe the 
story. He pointed out that its truth was hardly con- 
sistent with the recall of the late consul-general of 
Russia before his intrigues had been completed or with 
the rarity of his successor’s visits to the pasha. 4 And in 
this matter, as in most cases where Egypt was con- 
cerned, Campbell’s information was better and his 
conclusions more accurate than those of our impulsive, 
intemperate ambassador. 6 

1 Campbell, October 9, 1837, and enc. (F.O. 78-320). 

2 E.g. memo, by Palmerston, July 19, 1838 (F.O. 96-19). 

3 Cf. Wood to Ponsonby, December 31, 1835 (F.O. 1 95-1 07). 

4 Campbell, March 21, 1838 (F.O. 78-342). 

5 The correspondence of the Russian consulate, in course of 
publication, will, I believe, be found to confirm this view. 



What external influence moved the pasha to his next 
step was not the sinister promptings of Russian diplo- 
matists, but the sincere and freely expressed sentiments 
of merchants, both French and British. I have already 
mentioned their regret that he was not to be allowed to 
add Baghdad to the number of his cities. 1 Such feel- 
ings might perhaps have been explained away as the 
ignorant desires of men trading under the burden of 
corrupt incompetence. But they were held not only by 
the European merchants of Baghdad but by those of 
Cairo and Alexandria as well, and by their partners 
and correspondents at London, Paris and Marseilles. 
The pasha might pursue a system of monopolies, but 
under his rule as nowhere else in the Levant strict 
order and severe justice were established. His was a 
government with which bargains might safely be made. 
French and English merchants, however their govern- 
ments might differ, were emphatically agreed in 
desiring that his rule should be maintained and per- 
petuated. Waghorn, for instance, the active transit 
agent for the Suez route, seems to have assured the 
pasha that Britain would recognise his independence. 2 
British merchants refused to quit Cairo and Alexandria 
when the British consul-general withdrew and when 
British forces were attacking Ibrahim in Syria; and 
when the troubles were all over in 1842, a committee 
was formed in London to strike a gold medal to com- 
memorate the protection which the pasha had “nobly 
afforded” to British residents in Egypt, 3 while our 
consul-general was highly embarrassed at being desired 
to present to him an address from the Bengal Chamber 
of Commerce, applauding the dignified and impressive 

1 Vide p. 1 26, supra. 

2 Campbell, April 16, 1838, and enc. (F.O. 78-342). Cf. 
passage marked for omission in confidential to Campbell, Tune o, 
1838 (F.O. 78-343). 

3 Hoskins, op. cit. p. 290. 


example which he had set to the nations of Christen- 
dom. 1 He must, I think, be excused for mistakenly 
supposing that the popular view could not remain 
without influence on a popular government. 

On May 25, 1838, Muhammad ’Ali therefore made a 
formal declaration of his intentions to the French and 
British consuls-general, and a little later to their 
Austrian and Russian colleagues. He told them he had 
resolved to declare his independence of the Sultan. For 
this he gave two reasons — his family’s future, and the 
maintenance of his reforms. Campbell reports him as 
saying “that he never can consent that all that which 
he has been toiling for, and all the useful and costly 
establishments founded by him at an enormous expense, 
such as his arsenals, his fleet, his steam vessels, his 
manufactories with European machines and with 
workmen either European or natives who have been 
educated by him at great expense in Europe, the 
numerous useful schools and literary institutions which 
he has established entirely on the European system, the 
mines which he has opened both of coal and iron in 
Syria, and the roads and canals made there and in 
Egypt — he cannot, he says, ever permit all those esta- 
blishments to revert to the Porte and be lost at his death, 
and that he should have the pang of feeling that all his 
labours should merely have been for the Porte which 
would allow them to go to ruin, whilst his own family 
and children would be exposed to want and perhaps 
even to be put to death”. 2 

This proposal met at once with the strongest dis- 
couragement from both France and Great Britain. 
Cochelet was instructed to say that both states “were 
resolved if necessary to use coercive measures to 
contain Mehemet ’Ali in his duty of vassal to his 

1 Barnett, September 30, 1841 (F.O. 78-451). 

2 Campbell, May 25, 1838 (F.O. 78-342). 



sovereign ”. 1 Campbell, with abundance of moral 
advice, was to express the English cabinet’s deep con- 
cern mingled with hopes of a more just and prudent 
decision . 2 But it was still hoped that the crisis might 
be relieved without any actual explosion. Just at this 
time it was feared that the Sultan’s fleet might appear 
off the Egyptian coast. Muhammad ’Ali declared that 
if it did he would attack and destroy it in person . 3 
Palmerston at once suggested that the Turkish fleet 
might cruise in company with the British Mediter- 
ranean squadron, and go wherever that might go. 
This, he thought, would reassure France and the 
pasha and demonstrate that the Turkish squadron was 
out for exercise and instruction only, while it would 
also evince a close union between Turkey and Great 
Britain . 4 * 

Meanwhile the pasha received his discouraging 
answers with considerable calm, merely saying that he 
could not forgo his design, but that he hoped “the 
great powers would take a more just and equitable 
determination in his favour ”. 6 At this moment his 
hopes seem to have centred on the possibility of coming 
to some agreement with the Porte so as to cut away the 
technical grounds on which the great powers were 
resisting his proposals. Money was undoubtedly more 
powerful at Constantinople than at any other European 
capital. He had already enquired of Medem, the 
Russian consul-general, and of Campbell what their 
respective courts would do in case he arranged with the 
Turks for their recognition of him as an hereditary or 
independent ruler. He had received little encourage- 

1 Campbell, July 17, 1838 (F.O. 78-343). 

2 To the same, July 7, 1838 (ibid.). 

3 The same, July 12, 1838 (ibid.). 

4 To Ponsonby, July 25, 1838 (F.O. 78-329 A). 

6 Campbell, August 11, 1838 (F.O. 78-343). 


ment from either. 1 But the French were more favour- 
able. Early in the following year, 1839, they were still 
anxious to induce the pasha to abstain from hostilities 
by holding out the hope of some “arrangement being 
made with the Sultan favourable to the future position 
of his descendants”. 2 But this proposal Palmerston was 
unwilling to accept, at all events unless the pasha 
would give up most of Syria. 3 

Much encouraged by the general European attitude, 
which promised strong protection in case of defeat, 
Sultan Mahmud resolved upon the war which he had 
been so long preparing. Russian agents seem to have 
played the chief part of provocation, expecting that the 
Turks would be well beaten and would then summon 
the Russians back to Constantinople. 4 In February 
Ponsonby learnt that Mahmud had determined to 
make war in the spring. 5 He was reported to have sent 
an order to the Grand Council of his ministers, saying 
that the sar-’askar, Hafiz Pasha, reported his army 
able to beat Muhammad ’Ali’s, that the Capitan Pasha 
declared the fleet able to destroy the Egyptian fleet, 
and that it therefore remained for the Council courage- 
ously to do its duty. 6 Hafiz indeed and his German 
officers were incessant in their demands to march 
against Ibrahim in Syria. 7 In April therefore the Turks 
crossed the Euphrates at Bir. For two months nothing 
happened. Russia promptly demanded that Ibrahim 
should retire towards Damascus, promising in that 
event to induce the Sultan to withdraw from the 

1 Medem, March 20/April 1, 1838; and Campbell, July 9, 1838 
(F.O. 78-343). 

2 Granville, February 15, 1839 (F.O. 27-580). 

3 Cf. to Beauvale, June 28, 1839 (F.O. 7-278). 

4 Ponsonby, January 27, 1839 (F.O. 78-354). 

5 The same, February 12, 1839 {ibid.). 

6 The same, March 7, 1839 {ibid.). 

7 The same, March 19, 1839 (F.O. 78-355). 


Syrian frontier. 1 The pasha’s reply was a declaration 
that Ibrahim should retire as soon as the Turks had re- 
passed the Euphrates, and that if the four great powers 
would guarantee him from attack and support his 
desire of an hereditary succession, he would withdraw 
some of his men from Syria altogether and be prepared 
to accept a definite agreement. 2 The French addressed 
an urgent demand to Mahmud, calling on him to avoid 
hostilities, and declaring that unless Hafiz recrossed the 
Euphrates he would be deemed the aggressor. 3 At the 
same time they called upon Muhammad ’Ali to with- 
draw as well. 4 * By the middle of June the pasha, tired 
of waiting for some acceptable proposal, while the 
Turkish commander sought to raise a rebellion in 
Ibrahim’s rear, announced that he must at last give his 
son liberty to act. 6 On June 24, therefore, two hours 
after daybreak Ibrahim attacked Hafiz’s camp at 
Nazib. The German officers abounded with reasons 
why the Turks should have won. 6 But the action was 
more of a rout than a battle. Ibrahim captured all the 
Turks’ cannon and baggage, and their army vanished. 

This brilliant success was immediately followed by 
two other pieces of good fortune. On July 1 was 
announced the death of Sultan Mahmud. 7 Disappoint- 
ment and anxiety had told heavily upon him. Some 
months earlier the master of his wardrobe had without 
his knowledge caused his dresses to be taken in so as 
not to hang too loosely on his shrunken form, 8 and he 

1 Campbell, May 7, 1839 (F.O. 78-373). 2 Ibid. 

8 Ponsonby, June 16, 1839 (F.O. 78-356). 

4 Campbell, June 16, 1839 (F.O. 78-374). 

5 The same, June 14, 1839 (ibid.). 

6 Ponsonby, July 8, 1839 (F.O. 78-356). 

7 Beauvale, July 11, 1839 (F.O. 7-281), says the death had 

occurred on the 29th and had been concealed for thirty-six hours. 

8 MacCarthy and Caratheodory, Maladie . . . de . . . Mahmoud II, 
p. 20. 


had followed the preparations for his attack upon 
Muhammad ’Ali with incessant anxiety, which, it was 
said, he sought to relieve with forbidden liquors. His 
obstinate hatred had made him a dangerous enemy, 
and the pasha had good cause to rejoice at his death. 
He was succeeded by his eldest son, Abdul Majid, a boy 
of sixteen, brought up in the harim, with a dwarf and 
two black eunuchs for his favourite companions. 1 
Although Mahmud’s obstinacy had been enlightened 
by but few sparks of intelligence, the counsels of the 
empire would evidently be enfeebled by his death un- 
less some external guidance could be found. 

And while Ibrahim’s victory and Mahmud’s death 
were still in all men’s mouths, the Turkish fleet ap- 
peared off Alexandria, not to bombard the city, but to 
join the pasha. It has commonly been supposed that 
this result had been brought about by bribery. But 
other reasons suffice by themselves to explain the con- 
duct of the commander, the Capitan Pasha. Ahmad 
Mushir, who held that office, had been ordered to sail 
for the Syrian coast in order to co-operate with the 
efforts of Hafiz to raise revolts against Muhammad 
’Ali, and for this purpose he had some 6000 troops 
aboard his fleet. 2 But when he had passed the Darda- 
nelles, he received new orders to proceed to Rhodes. 
This aroused his suspicions, and he then learnt from 
the captain who had brought the orders that at Rhodes 
he was to be dispossessed of his command and the fleet 
returned to Constantinople. He assembled his officers, 
told them he was convinced Khusrau meant to give the 
fleet up to the Russians, and that it would be better to 
join Muhammad ’Ali, with which they all agreed. 3 He 

1 Beauvale, July 10, 1839 (F.O. 7-281). 

2 Campbell, July n, 1839 (F.O. 78-374). 

3 The same, July 17, 1839 (F.O. 78-375). Cf. Ponsonby,July 8, 
1839 (F.O. 78-356). 


had always been an enemy of Khusrau. Mahmud’s 
death threatened to invest Khusrau with greater 
authority than ever. It was natural for Ahmad there- 
fore to proceed to Alexandria and propose to join 
Muhammad ’Ali in overthrowing the common enemy 
of both. What would have been an act of treason in a 
European state was in Turkish politics a mere act of 
prudent foresight. “I have never known a Turk”, 
wrote Campbell, “ . . .who was not in all his acts guided 
by his own interest, or by ambition of power and the 
desire to overthrow his personal opponent .” 1 

This defection seemed to place the game in Muham- 
mad ’Ali’s hands. There was no force to hinder 
Ibrahim from marching upon the Bosphorus by land, 
while the united fleets appeared before Constantinople. 
Ponsonby was sure that the forts at the Dardanelles 
would show no vigour in opposing them, and that a 
new Divan might be formed in which Muhammad 
’Ali’s friends would have a decided control . 2 But the 
pasha, in accordance with his general principles of 
moderation, does not seem to have wished to go so far. 
As soon as he had heard of Mahmud’s death, he ordered 
Ibrahim to suspend hostilities. The day after Ahmad 
had dropped anchor in the harbour of Alexandria, a 
messenger arrived with a letter from Khusrau, formally 
announcing Abdul Majid’s accession. It was gracious 
in tone. The Sultan pardoned the pasha’s conduct to- 
wards his deceased father, promised new honours and 
the hereditary government of Egypt and its dependen- 
cies, and engaged the pasha to promote the prosperity 
of the empire . 3 These were not terms with which 
Muhammad ’Ali was prepared to be content. But he 

1 Campbell to Ponsonby, July 16, 1839 (F.O. 78-375). 

2 Ponsonby, July 21, 1839 (F.O. 78-357). 

3 Campbell, July 11, 1839 (F.O. 78-374). Cf. Ponsonby, July 
2 and 3, 1839 (F.O. 78-356). 


was confident he could now obtain what he really 
wanted, the hereditary government of his existing 
possessions, and spoke publicly of proceeding to pay 
his personal homage to the young Sultan. 

But “Ottoman ministers are an abject, miserable 
sort of men ”. 1 Khusrau, essentially false, and incapable 
of honest dealing, had sent other letters to Egypt as 
well as his gracious one to the pasha. He wrote to four 
of the chief officers of the fleet exhorting them to seize 
the Capitan Pasha and bring him back to Constanti- 
nople. Muhammad ’Ali at once took up the challenge. 
He wrote to Khusrau calling upon him to resign his 
office since he could be trusted by neither the chief men 
nor the “nation ” in general , 2 and he sent out a circular 
to the pashas of the empire demanding their aid in 
removing this unworthy Grand Vazir, whose conduct 
had never benefited either the throne or the “nation”, 
and who had been the cause of all the evils that for 
years had afflicted the state . 3 And at Constantinople 
men’s hearts failed them for fear. The only plan that 
Khusrau could devise to save him from the impending 
danger was to comply with the demands that the pasha 
had made, for the hereditary government of all his 
territories. But just as this decision had been taken, the 
Internuncio received instructions from Metternich that 
transformed the situation. To the Austrian minister, as 
to Soult at Paris and Palmerston at London, the 
situation threatened the possibility of Russian inter- 
vention under the Treaty of Unkiar Skelessi. The 
Internuncio was therefore ordered to engage the repre- 
sentatives of France, Prussia, Russia and Great Britain 
to join with him in presenting a note to the Porte de- 
claring that an agreement between the five powers was 

1 Ponsonby, July 8, 1 839 (F.O. 78-356) . 

2 Enc. in Ponsonby, August 6, 1839 (F.O. 78-357). 

3 Enc. in Campbell, July 28, 1839 (F.O. 78-375). 




assured and that the Porte would do well to take no 
action without their concurrence. The note was signed 
the very day the instructions were received and it 
was delivered to Khusrau early next morning. 1 It 
gave the Vazir courage enough to reverse the decision 
which he had taken. By order of the ambassadors their 
note was communicated to Muhammad ’Ali on August 
6. He was deeply preoccupied and his expression 
betrayed the uneasiness which this new and unexpected 
change caused him. 2 

Ponsonby was delighted at this development. He 
suffered from a highly virulent form of Russophobia, 
and smelt Russian intrigue in everything that transpired. 
He was particularly convinced that the pasha was sold 
to Russian interests, and had inoculated Palmerston 
with his ideas. So early as 1836, a question of trading 
dues which Muhammad ’Ali had reduced for Russian 
goods but had been unwilling to lower for British com- 
modities, had been deemed to confirm reports “that 
there is a closer understanding between the Pasha and 
the Russian Government than either of the two parties 
have hitherto acknowledged”. 3 Now everything that 
the pasha did was ascribed to Russian influence. There 
was an understanding between him, Russia, and 
Persia. 4 Russia was visibly favouring his cause. 5 A new 
party had been formed to overthrow Khusrau with 
Russian aid. 6 In vain did Campbell use every argu- 
ment he could think of to dispel this illusion. Our 
sharp-nosed ambassador could not even detect any- 
thing queer in the Russian ambassador’s readily sign- 
ing the joint note when his great aim was supposed to 

1 Ponsonby, July 29, 1839 (F.O. 78-357). 

2 Campbell, August 7, 1839, and enc. (F.O. 78-375). 

3 To the same, November 22, 1836 (F.O. 78-281). 

4 Ponsonby to Campbell, July 5, 1839 (F.O. 78-375). 

6 Ponsonby, August 19, 1839 (F.O. 78-358). 

6 The same, August 21, 1839 (ibid.). 


be the maintenance of the old precarious situation in 
order to secure the accomplishment of Russian aims. 1 

The first thing to be done was to prevent the pasha 
from acting while Europe heavily deliberated. In this 
both France and England were agreed. The French 
consul-general warned Muhammad ’Ali that the 
French and British squadrons might be used for co- 
ercive measures. Palmerston wrote, “ The pasha must 
be well aware that he is not in a position which, either 
geographically or politically or with reference either to 
military or naval considerations, can enable him with 
impunity to set at defiance the governments of Europe 
and more especially the maritime powers”. 2 

It was significant that at this time the Foreign Office 
decided to recall Colonel Campbell. He had then been 
serving in Egypt since 1833. He had watched closely 
the progress of the pasha’s policy, both internal and 
external. He had travelled extensively in Egypt, in 
Syria and Crete. He was no blind admirer of the pasha, 
whose conduct on occasion he could criticise with due 
severity; but his conciliating manners, his persuasive 
address, his commanding presence, his sound, just 
judgment, had given him great influence with Muham- 
mad ’Ali, who regarded him as a personal friend. But 
he had forgotten his own interest. When the tide of 
diplomatic opinion had begun to run against the pasha, 
Campbell had sought to stem it. He had tried to dis- 
abuse Ponsonby of his Russian fantasy. He had pointed 
out with untimely truth that the Jews and Christians of 
Syria would suffer acutely if they were restored to the 
direct rule of the Sultan. 3 He had dared to suggest that 
the Turkish empire might be restored to progress and 
prosperity if Khusrau were removed from office and 

1 Ponsonby, August 20, 1839 (F.O. 78-358). 

2 To Campbell, September 13, 1839 (F.O. 78-372). 

3 Campbell, August 7, 1839 (F.O. 78-375). 



Muhammad ’Ali invited to co-operate in its reform. 1 
It was unbearable that a man who merely knew the 
people and conditions of Egypt and Syria, and who had 
only seen with his own eyes the good that had been 
done, should even hint his disagreement with the 
official view that Muhammad ’Ali’s reforms were sham ; 
and it was monstrous that, on learning of Muhammad 
’Ali’s demand that Khusrau should be removed from 
his office, he had not expressed the shocked surprise he 
would have felt on hearing that Lord Auckland in a 
fit of madness had demanded that Lord Palmerston 
should cease to be Secretary of State. 2 In September 
he was curtly informed that Palmerston intended to 
advise his recall 3 — a measure which the minister had 
been contemplating for a year. 4 * By an ironical chance 
the news of Keane’s occupation of Kabul could only 
be forwarded without delay to Malta by a steamer 
which the pasha placed at Colonel Campbell’s dis- 
posal. 6 

Colonel Hodges, Campbell’s successor, reached 
Alexandria in December, 1839. 6 He proved to be hot- 
tempered, blustering and quarrelsome. He began by 
quarrelling with the packet agent at Alexandria for 
charging him postage in accordance with instructions 
from the Postmaster-General. 7 He went on to confide 
in a most intemperate vice-consul who carried tales 
well calculated to inflame both the consul-general and 

1 Campbell to Ponsonby, August 6, 1839 (F.O. 78-375). 

2 Cf. despatch to Campbell, August 13, 1839 (F.O. 78-372). 
Palmerston does not, of course, draw the parallel I suggest, but it 
seems to represent his feelings. 

3 To Campbell, September 11, 1839 (ibid.). 

4 Palmerston’s note, October 26, 1838 (F.O. 78-344). 

8 Campbell, October 12, 1839 (F.O. 78-375). 

6 The same, December x8, 1839 (ibid.). 

7 Hodges (consular), January 23, and to Hodges, July 3, 1840 
(F.O. 78-407). 


the Foreign Minister. 1 His own conduct won him the 
cordial dislike of the whole consular body, and before 
the consulate-general was reopened in 1841, he was 
very wisely sent off to cool his temper in the less trying 
climate of Hamburg. 2 His services in the crisis of 1840 
seem to have been more than amply recognised by 
permission to accept the insignia of a Turkish general 
of division. 3 

Two other incidents, slight in themselves, may be 
here mentioned as illustrating the hasty and incon- 
siderate temper with which matters were being driven 
on. Hodges mentioned that the Swedish consul- 
general had defended Muhammad ’Ali’s detention of 
the surrendered Turkish fleet. Without waiting for 
more, Palmerston at once requested his recall. The 
Swedish government desired reasons. Palmerston de- 
manded of Hodges a few good damning circumstances; 
but none were to be had. 4 5 

The second incident occurred a little later in 1840. 
On May 5 a despatch was addressed to Hodges relating 
to the trial of certain Jews at Damascus. He was to 
represent the disgrace reflected by these “barbarous 
enormities” on a ruler who prided himself on pro- 
moting civilisation. 6 A further despatch dwells upon 
“the deep and general feeling of indignation ” which 
had arisen throughout the country. 6 Here again is a 
prompt, unhesitating readiness to believe the worst. 
The case was one of those in which the Jews were 

1 Barnett to Bidwell, September 20 and December 17, 1841 
(F.O. 78-451). 

2 To Hodges, June 25, 1841 (ibid.). 

3 To the same, February 18, 1841 (ibid.). 

4 Hodges, January 24, 1840 (F.O. 78-404). To Hodges, 
February 25 and March 25, 1 840 (F.O. 78-403) . Hodges, March 
21, 1840 (F.O. 78-404). 

5 To Hodges, May 5, 1840 (F.O. 78-403). 

6 To the same, May 30, 1840 (ibid.). 


accused of murdering a Christian in order to mingle 
his blood with the unleavened bread of the Passover. 
Belief in the existence of such practices was still as uni- 
versal in the Levant as it had been in medieval Europe. 
The accused men had been seized, tried by the ordinary 
processes, and condemned. Most unhappily, there was 
but too much reason for thinking that Sharif Pasha, 
Muhammad ’Ali’s governor at Damascus, had acted 
with propriety. He had followed the advice of the 
French consular agent. Worse than that, the English 
consul, Werry, not only considered the guilt of the 
accused proved by the proceedings on their trial, but 
testified that the prompt and suitable measures taken 
by Sharif Pasha had saved the Jews of Damascus from 
general pillage and massacre . 1 

The nervous irritability which Palmerston displayed 
in these incidents resulted from the difficulties which he 
found in bringing the general question at issue to a 
satisfactory conclusion. Matters had gone very contrary 
to his expectations. The anticipated difficulty had been 
to induce Russia to co-operate with the other powers 
and to prevent her from consolidating her position by 
separate action. But this proved to have been greatly 
exaggerated. The Emperor Nicholas was not eager to 
act under the Treaty of Unkiar Skelessi. The reap- 
pearance of Russians at Constantinople would have 
meant war with England and probably with France as 
well, and in the Near East the consolidation of Muham- 
mad ’Ali’s power — an object remote from the emperor’s 
wishes. Moreover a divergence of views had already 
appeared between Palmerston and Soult, the French 
Foreign Minister. The first wished to restore Syria to 
the Sultan, the second to leave it in the pasha’s posses- 
sion. If therefore the emperor supported Palmerston’s 
policy, instead of attempting to act alone, he would 
1 Hodges, June 18 and July 20, 1840, and enc. (F.O. 78-405). 


have a good prospect of breaking up, instead of 
cementing, Anglo-French co-operation. 1 He therefore 
sent Baron Brunnow on a special mission to London in 

This cleared away one difficulty, only to raise another. 
Palmerston was eager to carry with him France as well 
as Russia, if he could. But French foreign policy was 
now, as it had been ten years earlier, beset by many 
difficulties. The July Monarchy, like its predecessor, 
was never strong enough to ignore currents of popular 
opinion; and French sympathies were strongly in 
favour of Muhammad ’Ali. And, as always, policy was 
liable to distraction by diverse and sometimes con- 
flicting continental and colonial interests. Fear of the 
press made it difficult for Soult to withdraw from the 
position he had taken up. His ministry fell, on a purely 
domestic issue, at the end of February, 1840, and he 
was succeeded by Thiers. 

The new Foreign Minister pursued Soult’s policy but 
with rising bitterness against England. His first plan 
was to revive separate negotiations between the Porte 
and the pasha through Pontois, the French ambassador 
at Constantinople, in order to confront Great Britain 
and Russia with a settlement which they could find no 
pretext to reverse. 2 Probably as a result of these en- 
deavours, Khusrau was removed; and Muhammad 
’Ali promptly decided to send his secretary, Sami Bey, 
on a mission to Constantinople. His pretext was the 
offering of congratulations on the birth of the Sultan’s 
daughter, and the concession which was to be offered 
was the restoration of the Turkish fleet. 3 The immediate 
answer to this was the signature by Great Britain, 
Russia, Austria and Prussia of a treaty by which 

1 Cf. Mowat, ap. Camb. Hist, of Br. For. Policy, n, 1 72-1 73. 

2 Medem to Nesselrode, May 1/13, and May 22/June 4, 1840. 

3 Hodges, Nos. 50 and 53, June 17, 1840 (F.O. 7&-405). 


Muhammad ’Ali was to receive the pashaliq of Egypt 
on a hereditary tenure and southern Syria for life, 
provided he accepted the offer within ten days; if he 
delayed acceptance beyond ten days but accepted 
within twenty days, he was to receive Egypt alone ; if 
he refused, the four powers would blockade him; if he 
advanced on Constantinople, the powers would co- 
operate at the Sultan’s request in its defence; and 
article 4 re-established “the ancient rule of the Otto- 
man Empire”, that the Dardanelles should be closed 
to all foreign ships of war whenever the empire was in a 
state of peace. The treaty was signed on July 15, 1840. 
Palmerston had succeeded in his great desire to merge 
the Treaty of Unkiar Skelessi in some more general 
agreement, although he had failed in securing the co- 
operation of the French. 

The news provoked an outburst of indignation at 
Paris. The French press, the ministers, the king, talked 
as though war were imminent. But they knew, and 
Palmerston knew, that it was not. “France will not 
help him [the pasha] ”, Palmerston wrote, “. . .nor has 
she the means of doing so.” She had indeed fifteen sail 
in the Mediterranean, but it was almost the whole of 
her navy. She had an army of 60,000 men in Algiers 
who needed large reserves to be maintained to fill up 
the gaps caused by “fever and the Moors”. “How 
in this state of things could France wantonly engage in 
war with the great military powers of the conti- 
nent?. . .” 1 

Thiers’ next hope was that matters could be so spun 
out as to leave the issue undecided when winter would 
disperse the blockading squadrons and prevent the 
movement of troops, when he could hope to break the 
concert that had been formed and reassert the influence 
of France. With this object in view, he advised the 
1 To Hodges, July 18, 1840 (F.O. 78-403). 


pasha to reinforce his position, to remain on the de- 
fensive, to hold out. 1 It was the worst advice that 
could have been tendered. A sudden advance on Con- 
stantinople would perhaps have produced such an 
upheaval as would have enabled the pasha to get 
better terms. But to refuse the allies’ offers and 
attempt a mere passive resistance was inviting defeat. 
This also was foreseen by Palmerston. France, he said, 
“will wait, will lie by, and if Mehemet Ali should be 
able to resist the allies for any length of time, she will 
then offer herself as a mediator. But it will be the 
business of the four powers so to press Mehemet Ali as 
not to give France such an opportunity”. 2 

In the face of the formidable combination that had 
been slowly forming against him, the old pasha had 
continued to hold his head high. He probably found it 
impossible to believe that the great powers would really 
agree upon a common course of action in a matter over 
which they had always been sharply divided. He had 
reckoned on either France or Russia neutralising the 
action of Great Britain — if Great Britain did attempt to 
act. When Hodges was ordered to urge upon the 
Turkish naval officers their duty of rallying “round 
their sultan and caliph”, 3 the pasha leapt up from his 
divan vowing he would shoot the first deserter, and 
Hodges decided that he had better not carry his orders 
into effect. 4 New regiments were raised; troops were 
recalled from Arabia; a camp of 36,000 men was 
formed at Damanhur, a judicious choice from its 
central position; and all with a method and good order 
that Hodges had not expected. 5 Indeed even Hodges 

1 Instructions to Cochelet, July 29, 1840. Sabry, op. cit. p. 501. 

2 To Hodges, ut supra. Cf. Palmerston to Hobhouse, July 27, 
1843 (Brit* Mus. Add. MSS, 36471, ff. 21 1 sqq .). 

3 To the same, February 25, 1840 (F.O. 78-403). 

4 Hodges, March 31, 1840 (F.O. 78-404). 

5 The same, February 21, 1840 {ibid.). 


1 86 

began to be impressed with this old man’s life, vigour 
and intelligence, and to fear lest in despair he should 
cause some general conflagration “whence might 
spring new interests, new combinations, new chances 
in his favour”. 1 But by the middle of the year the 
strain of unceasing anxiety had begun to tell. He was 
shaken by “feverish and restless starts”. 2 In August 
the Russian consul-general found him asleep on the 
divan. He said he had been unable to sleep for several 
nights. “His Highness’s state of health, the torments to 
which he was evidently a prey, the efforts he made to 
control the irritation caused by his present position, 
and the conflicting feelings that possess this old man of 
70 and combine to wear out the energy which dis- 
tinguishes him, made our conversation painful in- 
deed.” 3 

But he lost neither his grasp of affairs nor his skill 
in counting chances. Like Thiers, he reckoned the 
allies would be slow to act, and that their blockade, 
when formed, would accomplish little immediately. 4 
He saw he could rely upon the moral, though not upon 
the material support of the French. He knew that 
English public sentiment was far more friendly to him 
than to the Porte. He reckoned therefore that if matters 
went ill he could at least rely upon the hereditary 
government of Egypt, while, if the alliance by some 
lucky chance broke up, he might get all Syria too. 6 
When therefore an agent from Constantinople and the 
consuls-general presented the allied demands, he re- 
fused to listen to the pompous eloquence 6 of Colonel 

1 Hodges, July 26, 1840 (F.O. 78-405). 

2 The same, July 5, 1840 (ibid.). 

3 Medem to Nesselrode, August 19/31, 1840. 

4 Walewski’s report ap. Sabry, op. cit. p. 508. 

6 Idem, p. 509. 

6 He liked to convey Palmerston’s views “impressively”, e.g. 
despatch of January 4, 1840 (F.O. 78-404). 


Hodges and desired a written statement. 1 The ten days 
passed without any formal answer. When the twenty 
days were almost expired, he offered to accept the 
second alternative but refused to confirm his agree- 
ment by the immediate release of the Turkish fleet. 2 
The term expired, but still the consuls-general lingered 
on at Alexandria, although news arrived on September 
7 that the Sultan had removed Muhammad ’Ali from 
all his offices and that the consuls-general had been 
recalled. 3 They did not actually quit Alexandria until 
September 23. 4 

One reason of this delay seems to have been their 
desire to watch the conduct of the French consul- 
general. Another was the fact that their own mutual 
confidence was slender. For instance on September 7 
a steamer put into Alexandria harbour from Beyrout, 
and sent off about £5000 in Turkish money under the 
English flag to be put on board an English man-of-war 
lying in the harbour. The boat and money were seized 
by the harbour-master, because according to the 
Turkish regulations the currency of the country might 
not be exported. Hodges in great anger threatened to 
haul down his flag at once. To the Russian and 
Austrian representatives this seemed dangerously likely 
to provoke a direct quarrel between the pasha and 
Great Britain, which would have given the latter an 
opportunity of acting separately on her own account. 
They therefore intervened to smother the business. 5 

This incident, provoking as it must have been, was 
not, however, the greatest humiliation that Hodges 
suffered during his last few days at Alexandria. There 

1 Hodges, August 20, 1840 (F.O. 78-406). 

2 The same, August 30, 1840 (ibid.). 

3 The same, September 15, 1840 (ibid.). 

4 Medem to Nesselrode, September 13/25, 1840. 

5 Same to same, September 2/14, 1840. 


was the question of the Indian mails. Months earlier 
Hodges had been ordered to ascertain the pasha’s in- 
tentions in this matter should coercive measures be 
adopted against him. 1 On September 19 a packet 
arrived. Hodges, not knowing what to do, visited the 
Divan, and expressed a hope that the mails would not 
be molested. The pasha merely nodded. The consul- 
general asked assurances. The other replied that he 
would give none. Hodges said he was much surprised. 
“The powers calling themselves civilised”, retorted the 
pasha sharply, “have adopted measures which may 
perhaps force me to imitate their example”; and when 
asked to explain himself added “their declarations are 
not to be depended upon”. Hodges could not accept 
this remark if it applied to Great Britain. With a 
sardonic laugh Muhammad ’Ali answered, “You may 
take it or leave it; but my remark is in the mouth of 
everybody”. Finally he said the mails might pass for 
this time only. So, very hot and angry, Hodges went 
back to his consulate to inform Lord Palmerston and 
the Bombay Government that the mails could not pass 
in future. 2 The same evening, in conversation with Her 
Majesty’s Postmaster, Hodges was told that “ some 
person [the italics are his own] had raised false and 
unnecessary alarm” about the mails. 3 Next day he 
heard from the Russian consul-general that Muham- 
mad ’Ali had assured the East India Company’s agent 
that so long as he ruled Egypt the mails should pass in 
perfect safety. 4 On this the consul-general’s wrath 
boiled over. He sent home an indignant complaint 
against the Postmaster and the Company’s agent. “The 
question comes to be whether Mehemet ’Ali is to make 

1 To Hodges, April 14, 1840 (F.O. 78-403). 

2 Hodges, September 20, 1840 (F.O. 78-406). 

3 The same, private, September 22, 1840 {ibid.). 

4 The same, September 22, 1840 {ibid.). 


a disaffected functionary a stalking-horse in order to 
deride Her Majesty’s Agent, to depreciate his con- 
sideration, to erect an anomalous English authority in 
his stead, and in a word to reduce that agent to a 
political nullity.” He had expected, he says, nothing 
but hostility from other Englishmen there, but hoped 
that persons in public employment “would have 
sympathised with me”. 1 Palmerston at least sym- 
pathised so far as to complain to the President of the 
Board of Control against the Company’s agent, only to 
learn that “if the complaint is communicated to the 
chairman, it will be made known to the court and con- 
sequently become public”. 2 

Muhammad ’Ali was as good as his word. In spite of 
the withdrawal of the consul-general, the hostilities that 
took place in Syria, and the disorders that threatened 
to break out in Egypt, he not only suffered the mails to 
pass, but took special measures to protect travellers 
using the Suez route. 3 His war, he said, was with Lord 
Palmerston, not with the English. 

But although he certainly had the best of the joke, 
he could not but get the worst of the match. The forces 
against him were too strong, their direction too prompt 
and vigorous. On September 11 a force of British 
marines and Turkish troops were landed on the Syrian 
coast near Beyrout. For months a general unrest and 
Ottoman emissaries had been stirring up the Syrians to 
revolt. Ibrahim’s army was scattered, weak, lacking 
stores and supplies. In October the Druses rose. On 
October ioatBait-hannis, Commodore Napier met and 
defeated Ibrahim in person at the head of a small body 
of troops, and captured his standard. Beyrout was 
taken. On November 4 Acre, which had resisted 

1 Hodges, private, ut supra. 

2 Hobhouse to Palmerston, October 9, 1840 (F.O. 78-451). 

3 Parbury, Handbook to the Overland Route, p. 257. 



Ibrahim for six months, surrendered after a single day’s 
bombardment. The Egyptian power in Syria had 
collapsed. At Paris, the Thiers ministry which had 
dragged France disagreeably near to war, had fallen a 
few days earlier, on October 29. On November 15 
Commodore Napier appeared off Alexandria with a 
strong squadron. On the 27th, though invested with no 
diplomatic authority, he signed a convention with the 
pasha, who agreed to evacuate Syria, and to restore the 
Ottoman fleet, on the condition that he should be 
recognised as hereditary governor of Egypt. Orders 
recalling Ibrahim from Syria were issued on the 29th. 1 

This news took the diplomats of Constantinople 
completely by surprise. “Old Napier”, wrote Hodges 
dropping for the nonce his consular dignity, “has 
kicked up a devil of a row among the corps diplomatique 
here.” 2 He had indeed behaved with nautical irregu- 
larity in communicating to the pasha the decision 
which Palmerston and the English cabinet had taken 
on October 10, in deference to French feeling. This was 
to recommend Muhammad ’Ali as hereditary pasha in 
Egypt provided he speedily withdrew his troops from 
the other Turkish provinces and gave up the Turkish 
fleet. 3 When Napier’s agreement reached London, it 
was promptly approved. Ponsonby’s prejudices still 
invented obstacles in the way of a full settlement. 
He induced the Porte to issue a farman — the hatti 
sharif of February 13, 1841 — containing a number of 
unwelcome restrictions. 4 This, on Napier’s advice, 
Muhammad ’Ali rejected. Both Palmerston and 
Metternich urged the modification of the grant. This was 
done by a farman of June 1. It recognised a hereditary 

1 Letter of Shawal 3, 1256 (Abdine Archives). 

2 Hodges to Bidwell, January 21, 1841 (F.O. 78-451). 

3 To Ponsonby, October 15, 1840 (F.O. 78-390). 

4 Sabry, op. cit. pp. 532-533. 


right by seniority, 1 in the direct male descendants of the 
pasha. It fixed the tribute at 80,000 purses of piastres. 
It limited the pasha’s army to 18,000 men save in time 
of war or by express permission. It forbade the building 
of new ships. The ruler of Egypt was thus never to be 
able to threaten again the peace of Europe. Muham- 
mad ’Ali had failed to found his empire. But he had 
secured much. Egypt was virtually independent of the 
Porte’s control. Its administration was separate. This 
privilege was guaranteed by the assent of the great 
powers, and though the pasha had failed in his larger 
scheme, he had in fact laid the foundations of a new 

1 According to the Turkish rule of succession. 

Chapter VII 


I have already observed that among the causes leading 
Palmerston to oppose the extension of Muhammad 
’Ali’s power must be reckoned his misunderstanding of 
the pasha’s administration. It was indeed the object 
among contemporaries of immoderate praise and un- 
measured blame. Some, like the enthusiastic Waghorn, 
would only see the good that had been done and would 
not confess that many of the pasha’s achievements were 
“showy rather than substantial”; while others, like 
Dr Holroyd, Palmerston’s correspondent, viewed 
matters with English eyes and commiserated the 
fellahin because they did not live in brick cottages and 
eat beef . 1 But in order fairly to judge the pasha’s 
reforms and administrative ideas, a number of points 
must be borne constantly in mind. One is that he was 
working in an oriental country, where the functions of 
government were poles asunder from the customs pre- 
valent in the West. It was very hard for the zealots of 
laissez-faire to appreciate a system which directed the 
subject on every point. The parallels of Indian Govern- 
ment were of little use; for in those times few in England 
outside the India House and the Oriental Club had any 
sound understanding ofwhat their own countrymen were 
doing in India, so it is small wonder that the work of 
Muhammad ’Ali was misunderstood. Indeed little sym- 
pathetic intelligent criticism was to be got except from 
men such as Salt or Campbell, who knew the country 
well, or from Anglo-Indians who were familiar with 
similar problems and an apathetic eastern population. 

1 Campbell, December i, 1837 (F.O. 78-322). 


Then, too, the pasha had taken over a derelict 
government. It would be difficult to exaggerate the 
wretched condition of the Turkish provinces at the 
beginning of the nineteenth century. An honest 
governor, as Burckhardt said with perfect truth, could 
not hope long to hold his office. “The Porte demands 
supplies and nothing but supplies; and the pasha to 
satisfy her must press upon the industry of his subjects. 
He who is the well-wisher of his people, who contents 
himself with the ordinary revenue, and who lets justice 
preside in his councils will undoubtedly incur his 
sovereign’s displeasure, not because he is just, but 
because his justice prevents him from plundering and 
transmitting a portion of the acquired plunder to the 
divan. To save his existence he had nothing left but 
silently to resign his unhappy subjects to the rod of a 
succeeding despot or to declare himself a rebel, and to 
contend with his rival until the Porte, convinced of the 
difficulty of deposing him, patiently waits for a more 
favourable opportunity.” 1 These words were written in 
1810, but they are curiously prophetic of Muhammad 
’Ali’s career. Failure to understand this was to lead 
Palmerston to misjudge the pasha’s work completely. 

Egypt was probably in the worst condition of all the 
Turkish provinces, with the possible exception of ’Iraq. 
The Mamelukes had done nothing but oppress it. They 
protected the fellah neither from the arms of the 
Bedouin, nor from the extortions of the tax-gatherers, 
nor even from the ruin of his lands by the silting up of 
the canals. The delta, the most fertile land in the world, 
had lost a third of its cultivable extent. Fayum was 
dispeopled by Bedouin raids. Nobody knew what was 
exacted from the peasant or how much of the public 
revenue was pilfered on its way to the treasury. The 
revenue farmers — roznamji was their local name — were 
1 Burckhardt, Nubia , p. xxxviii. 

DM 13 


proverbial for their pride and wealth. 1 Justice was a 
matter of bribes, property a matter of favour, life a 
matter of luck. 

This was the kind of government which Muhammad 
’Ali inherited, to which he was accustomed, under 
which in Albania he had grown up. The establishment 
of his authority made him an absolute despot. None 
can wonder that he accepted his inheritance and on 
occasion acted much as his predecessors would have 
done. Jabarti, the historian, returning from the 
Shubra Palace to Cairo one June night in 1822, was 
strangled and his body tied to the feet of his ass. Men 
said that the pasha had been annoyed at the freedom of 
his comments. 2 The bringer of a secret letter is reported 
to have been thrown into the Nile to make sure he kept 
the secret if he had it. 3 The pasha beat a Copt until he 
consented to deliver over half his savings. 4 Even at the 
end of his days the autocrat ran strong within him. The 
wealthier people of Alexandria disliked sending their 
sons to Paris to be educated, and substituted the sons of 
porters and other mean persons for the children whom 
Muhammad ’Ali demanded of them. “ If these fellows ”, 
he said on learning of their conduct, “will neither 
understand the advantages of education nor of com- 
merce, they are only fit to carry loads on their backs 
like porters or donkeys.” Accordingly an order was 
made that all classes were to work in person on re- 
moving the mounds of rubbish which surrounded the 
city. Shopkeepers, merchants, scribes and theologians, 
were to be seen on the appointed days with baskets of 
earth on their backs and wholly unaccustomed sweat 
running down their faces. 5 

1 Paton, Revolutions in Egypt, 1, 79. 2 Jabarti, op. cit. 1, p. ix. 

3 Senior, Conversations in Egypt, 11, 116. 

4 Burckhardt, Arabia, 1, 442 n. 

5 Murray, July 8, 1847 (F.O. 78-708). 


And he was an Eastern despot not only by inherit- 
ance — he was one by environment as well. Apart from 
the small, weak European element — consisting of the 
consuls-general, a few French and English merchants, 
and the few Frenchmen in his service — he lived among 
people who expected and desired nothing else. The 
despot is always a lonely person. But Muhammad ’Ali 
was separated from his fellows not only by his unlimited 
power but also by his policy and intentions. “Do not 
judge me by the standards of your knowledge. Com- 
pare me with the ignorance that is around me”, he 
once said to Dr Bowring, who had been sent to report 
on the trade of Egypt and Syria. “ ... You have numbers 
of intelligent persons . . . ; I can find very few to under- 
stand me and do my bidding. I am often deceived, and 
I know that I am deceived. I have been almost alone 
for the greater part of my life .” 1 The good in the pasha’s 
rule was his own; the bad was generally the work of 
those whom he was obliged to use for want of better 
men — unscrupulous officials eager for money . 2 “When 
I came to Egypt”, he said again, “it was really bar- 
barous, utterly barbarous. Barbarous it remains to this 
day. Still I hope that my labours have rendered its 
condition somewhat better than it was. You must not 
however be shocked if you do not find in these countries 
the civilization which prevails in Europe .” 3 

It is certain that thirty years of his rule produced an 
extraordinary change. Yet a generation is too short a 
time to produce permanent results. The fact that the 
pasha found few or none to adopt his views and purposes 
with a sincere enthusiasm, and his consciousness of the 
gulf between his own policy and other men’s, in itself 
led to elements of weakness and instability in his work. 

1 Bowring, Report {Pari. Papers, 1840, xxi, 146). 

2 Campbell to Bidwell, December 1, 1837 (F.O. 78—322). 

3 Hodges, June 18, 1840 (F.O. 78-405). 



He felt — and justly — that every improvement depended 
on himself alone, and that what he left undone would 
perhaps never be accomplished. Hence arose in some 
of his undertakings a lack of due consideration at the 
outset, and a strong impatience to see immediate 
results. While he should have been busied with laying 
deep foundations, he was hurriedly raising the walls of 
the palace of his dreams. “I am old”, he would say, 
“. . .and what I would have done must be done quickly.” 

Many influences thus conspired to impair his re- 
forms, to rob them of a permanent driving-force, or to 
misdirect his activities. Yet it is hard to call to mind an 
oriental ruler who, under no spur of external necessity 
but solely from his love for order and justice and well- 
being, and in spite of the obstinate though passive 
resistance of almost all around him, succeeded in 
establishing a greater number of improvements. 

In the form of government he made small change. 
It continued along the lines which an age-long ex- 
perience had shown to be most appropriate, and which 
we ourselves hardly ventured to begin to alter in India 
until a generation ago. The village was the unit of 
administration, with its head-man, the shaikh-al-belad, 
representing the ruler in every capacity. The villages 
were grouped in subdivisions under the hakim-al-khot. 
These subdivisions were formed into sixty-one districts, 
each under a mamur , corresponding with the Indian 
collector. The districts were grouped in seven provinces 
each under a mudir, or governor, the area of whose 
jurisdiction was formed by an amalgamation of the 
twenty-four provinces into which the country had been 
divided in the time of the Mamelukes. In the larger 
cities some more elaborate organisation was necessary. 
There were special police and judges to maintain 
public order and to prevent or punish crime. And there 
too the population was divided according to trades and 



occupations into guilds, each under its special shaikh 
or head-man. In Cairo, for instance, there were 164 
of these bodies, the shaikhs being responsible for the 
conduct of the members of their respective guilds. 1 
This again was the traditional mode of organisation, 
prevalent throughout the East. 

In order to keep it working with a moderate degree 
of honesty and justice, a close, active and perpetual 
supervision was needed, the more so because popular 
opinion accepted official dishonesty as a thing of course. 
The ideal of the system was to maintain one tyrant to 
prevent the appearance of many. Left to themselves, 
the village shaikhs seized every occasion of oppressing 
their fellow-peasants. 2 Mamur and mudir oppressed 
all within their reach. Nor was dishonesty their only 
vice. They were ignorant as well. The studious might 
be deeply read in the philosophy of Islam and the poets 
of Arabia and Persia, but their schools bred scholars 
only, not men of affairs. Experience was the adminis- 
trator’s only guide, and that too often only showed the 
way to robbing with decency and prudence. 3 The need, 
too, of checking dishonest combinations among the 
officials demanded their frequent change of office, so 
that men were perpetually in control of matters of 
which they knew little or nothing. “Little attention is 
paid”, Bowring observes, “to the fitness of an indivi- 
dual for exercising the functions with which he is 
invested.” 4 Unfortunately this was inevitable. Camp- 
bell, no unfavourable witness, bears strong testimony 
to the point. “The vexations to our merchants”, he 
says, “arise in most cases from a want of system and 
from ignorance of business on the part of the local and 

1 Bowring, Report {Pari. Papers, 1840, xxi, 117). 

2 Gf. Burckhardt, Arabia, 1, 145. 

3 Puckler-Muskau, op. cit. 1, 24. 

4 Report {Pari. Papers, 1840, xxi, 49). 


minor authorities, rather than from any fault of 
Mehemet Ali or Ibrahim Pasha, and this arises chiefly 
from the absolute want of people as yet capable of con- 
ducting affairs in the numerous branches of administra- 
tion and in the many posts and places where European 
commerce extends, and there are also so many in- 
herent vices and venality in everything Turkish that it 
must be the work of time to prevent many evils which 
are at present unavoidable, but which daily diminish.” 1 
Indeed no permanent improvement was possible until 
a new generation had arisen, better taught and more 

By exhortation, punishment and inspection the pasha 
strove hard to make good the defects of training, 
education and character. His exhortations, conveyed 
in the form of circular letters, make queer but instruc- 
tive reading, and range from the ludicrous to the 
pathetic. Sometimes they contained awful threats. 
One, issued in 1826, complaining that his officials 
were not taking pains enough in promoting cultivation, 
declares that he is about to make a personal inspection, 
and in whatsoever district he finds any trace of negli- 
gence he will dig a pit and bury alive all the officials. 2 
But such warnings cannot have been meant seriously, 
for a little over a year afterwards, treating of the same 
kind of trouble, he merely says that his further advice 
will be delivered with the cudgel or the sword. 3 In 
others, which the consul-general Barker spitefully sent 
home to amuse Palmerston, he runs riot in abuse. 
Neglect to pay in the taxes affords the occasion of one. 
“From this it clearly appears that you are a negligent 
blockhead and affords another proof that you are an 
ass.” If the money is not immediately paid in, “be 

1 Campbell, December 12, 1838 (F.O. 78-343). 

2 Circular, Jumadi-us-sani 13, 1241 (Abdine Archives). 

3 Circular, Jumadi-al-awal, 1243 (ibid.). 



assured I will tear thee in pieces”. 1 On a failure to 
provide the requisite number of conscripts, he writes : 
“And thou — ass that thou art ! — what art thou doing? 
. . . For want of better men I have placed thee in 
office and made thee a governor, and thou ! dost thou 
neglect thy duty in this way and delay it so long?. . . 
In the instant thou receivest this order, put thy brains 
in thy head and send the rest of the men. . ..If thou 
dost not, I will make an example of thee to the rest of 
the governors of the districts . . . .” 2 In milder vein is an 
exhortation addressed to the Governor of the Sudan 
who had sent up a bag of rebels’ ears as proof of his 
activity. “Those in power and authority should know 
that the conquest of the country is to be secured by 
peaceful persuasion and the awakening of the natives’ 
confidence by means of justice.” The governor should 
follow the example which the French had set in Egypt, 
and imitate the behaviour of the English after them. 3 But 
a circular of 1843 exhibits the nearest approach to any- 
thing like an administrative testament. The old pasha, 
as he had then become, calls upon the zealous help of 
his servants, since his difficulties were far too great to be 
overcome by any single man. He reminds them of the 
fertility and situation of Egypt. “To possess a land like 
ours which has no equal is our great happiness, and to 
spare any effort by which her prosperity may be in- 
creased would be an act of grievous ingratitude which 
my heart cannot accept and in which I will never 
acquiesce. I must then ever and severely be summon- 
ing you to do your duty so that we may reach the end 
we have set before us. . . .Beware of idleness and negli- 
gence The wise man boasts not of his own conduct, 

1 Enc. in Barker, February 19, 1833 (F.O. 78-231). 

2 The same, January 23, 1833 {ibid.). 

8 To the Commandant in the Sudan, Rabi-al-awal 9, 1236 
(Abdine Archives). 


but of the success of the affairs entrusted to him. Know 
that I will pursue the well-being of this land even 
at the cost of my life and the lives of my kindred. All 
about me well know that I love not to harm any man. 
For forty years I have held my hand from sharp 
punishment, but if I am compelled to do otherwise, the 
fault will lie not at my door. . . . Of old I did not hope to 
reach the state to which to-day we have arrived ; and 
now that my ambitions are higher, I will readily 
sacrifice to the prosperity of my country, which is my 
great desire, even one of my own kin set over three 
million men.” 1 Three months later he made all his 
chief officials swear to serve him honestly and to report 
any abuse of authority that came within their know- 
ledge. 2 

There is no doubt that the circular just noticed marks 
the real wishes of the pasha’s heart. It was addressed 
exclusively to his officials. It was never communicated 
to the consuls-general. It was not designed to impress 
European opinion. It agrees with the language which 
Muhammad ’Ali used when speaking in confidence 
with European friends. But he was well aware that 
punishment was necessary as well as exhortation. It 
was true that he did not love to injure anyone, and in 
general withheld his hand from chastisement. But it 
was no more than a general disposition from which at 
times he could not help departing. For instance, in 
1822 the revenue collector at Ghiza falsely represented 
that he could get in neither the tithe nor the tax on 
houses. This was a most serious matter in the pasha’s 
eyes. He believed (it is now impossible to say whether 
justly or not) that the man was lying, probably from 
motives of corruption. He therefore instructed Ibrahim 
Pasha, at that moment acting as mudir of the province, 

1 Circular, Jumadi-us-sani 4, 1259 (Abdine Archives). 

2 Barnett, October 16, 1843 (F.O. 78-541). 



to reason with the man and if possible convince him of 
his error. If he succeeded, so much the better ; if not, 
the collector was to be beheaded, lest the interests of 
government should suffer by his misconduct. Ibrahim 
apparently carried out the sentence with his own hand. 
A later letter declares that the man had been slain by 
his own obstinacy, not by either the pasha or his son, 
and his place was to be filled either by a Frenchman or 
by the deceased’s brother. 1 As time passed, and 
perhaps as the general standard of conduct improved, 
the death penalty was used more and more sparingly, 
and all but great misconduct was punished by labour 
on the public works. An order of 1830 directed that 
twenty-five officials of middle Egypt were to undergo 
six months’ hard labour. 2 In 1833 the mamurs were 
warned that they would be thus punished if they com- 
pelled the servants of government to till land within 
their jurisdiction. 3 In the following year, as the spilling 
of blood was an act evil in itself, the mudirs and mamurs 
were forbidden to inflict the death penalty without the 
special orders of the pasha, 4 and in 1836 a village 
shaikh was ordered to be executed if he proved to be 
guilty of beating a man to death without provocation. 5 

But the best security against official misconduct lay 
neither in sound advice nor in bloody punishment, but 
in frequent and exact inspection. In this matter the 
pasha did not spare himself. He frequently went on 
tour, enquiring with the utmost particularity into the 
state of the accounts and the general administration ; 
and at such times, travelling virtually without a guard, 
was accessible to the complaints of the humblest fellah. 

1 To Ibrahim Pasha, Shaban 5 and 15, 1 237 (Abdine Archives). 

2 To Kotkhuda Bey, Safar 5, 1246 (ibid.). 

3 To the mudirs, Muharram 9, 1249 (ibid.). 

4 To the same, Rabi-us-sani 20, 1250 (ibid.). 

* To the mudir, Tanta, Jumadi-us-sani 6, 1252 (ibid.). 


One result of his tour in the Sudan in 1 839 was an 
order for the replacement of corrupt and ignorant 
officials. 1 

The number of Europeans employed in general ad- 
ministrative business seems to have been exceedingly 
small. There were in the country a number of French 
and English, renegades and others, but, while these 
men were freely employed in the arsenals, and in the 
army, 2 I think they were seldom set to administrative 
work, and the only definite reference to their civil 
employment I have met with is the one noted above, 
when Ibrahim was told he might appoint a Frenchman 
at Ghiza instead of the executed Copt. 

Nor were natives of the country employed in any of 
the superior posts. The superior administration was 
Turkish, not Arab. “The meanest man who speaks 
Turkish”, observed Bowring, “is ipso facto considered 
as belonging to a caste high above the indigenous in- 
habitant.” 3 An Arab servant could hardly even be 
sent with a message to a high official. The Turk in 
Egypt, even under Muhammad ’Ali, enjoyed something 
of the position of unquestioned superiority that the con- 
temporary servant of the East India Company did in 
India. Foreigners noted with surprise the universal 
sentiment of inferiority and subjection prevalent among 
the natives of the land. “We are but fellahin”, they 
would say. They never dreamt of questioning the right 
of the foreigner to rule over them. They were com- 
pletely unarmed, completely submissive, asking for 
nothing but to be allowed to pour Nile water over their 
fertile lands in peace. 4 

But the pasha was unwilling to allow this state of 
affairs to continue. He did not trust the Turks very far. 

1 To Abbas, Muharram 11, 1255 (Abdine Archives). 

2 Campbell, June 12, 1837 (F.O. 78-319). 

3 Report {Pari. Papers, 1840, xxi, 7). 4 Ibid. 


They were likely to sympathise with Constantinople and 
to yearn for the old, easy, corrupt and profitable ways 
of government which he was bent on reforming. He 
therefore sought as far as possible to replace them by 
Arabs. This bold idea, as it seemed at the time, was 
first suggested to him by Drovetti, the French consul- 
general. So early as 1826 no less than forty-five young 
men, the sons of Arab shaikhs and others, were sent to 
France to be educated there and rendered less unfit for 
public employment. 1 In 1833, in the course of a tour 
in the Delta, he visited the lower Turkish revenue 
officials with the bastinado, finding they had small 
sympathy with the Arab population and extorted 
money for their private use, and he declared besides 
that in future the Arab shaikhs should communicate 
directly with himself. 2 One result of this decision was a 
gathering of the principal shaikhs at Alexandria a few 
months later. Information given by Muhammad ’Ali’s 
secretary to the consuls-general suggests that this was 
done in order to permit the pasha to instruct and exhort 
them on the due discharge of their duties. Campbell 
reported an amiable dialogue in which the shaikhs 
promised the most scrupulous observance of the pasha’s 
orders. 3 But the published information does not seem 
to have told everything. Muhammad ’Ali had found 
that he could not prudently go too far in the employ- 
ment of Arabs. The Turks, as a foreign observer said, 
“always stole more decently than the Arabs”. 4 Besides 
that, [intriguers, who always flourish under a personal 
despot, however benevolent, were seeking to exploit 
the pasha’s evident good will. Influential village 
shaikhs, the pasha learnt, were stirring up their 

1 Salt, April 4, 1826 (F.O. 78-147). 

2 Campbell, June 13, 1833 (F.O. 78-227). 

3 The same, October 26, 1833 (F.O. 78-228). 

4 Puckler-Muskau, op. cit. 1, 24. 


brethren to delay the collection of the taxes in the hope 
that the blame would fall on the Turkish officials and 
that the shaikhs would be appointed to fill their places. 
This was to be brought to an immediate end. No time 
was to be wasted on tedious and inconclusive enquiries 
which would elicit nothing but lies. Any important 
village shaikh accused of acting thus was to be punished 
without more ado . 1 It seems unlikely that this business 
was forgotten in the shaikhs’ assembly at Alexandria, 
although the report given to Campbell says nothing of 
it. Further development of Arab employment had to 
await the slow development of the pasha’s educational 

At headquarters the work was divided among seven 
departments — war, the navy, cultivation, finance, 
commerce (and external relations) education, and 
police; but though the ministers in charge of them 
were reckoned of higher rank than the mudirs of the 
provinces, they seem to have had no separate authority 
of their own over the latter, for the pasha was careful 
to hold all the strings of government in his own hands. 
Nor did he suffer these central departments to follow 
their universally instinctive course and swell into vast 
establishments whose main purpose is to justify their 
employment by the complication of public procedure. 
He suppressed 200 posts in the Treasury, and, not con- 
tent with that, reminded the director that the chief 
merchants of Alexandria could control with only four 
clerks a business turnover amounting to nearly a third 
of the Treasury receipts. Had he forgotten the chief 
accountant’s mania for filling the public offices with 
Copts? If he could not manage better, the directorship 
would be given to someone else . 2 

* To Abbas Pasha, Jumadi-al-awal 28, 1249 (Abdine Ar- 

2 To Sharif Pasha, Rabi-us-sani 26, 1260 (ibid.). 



The most interesting aspect of Muhammad ’Ali’s 
rule is certainly the pains he took to develop and en- 
large the practice of discussing public business before 
proceeding to any action. In 1819 he set up a council 
or divan of seven persons to manage and discuss 
transactions between the Treasury and European 
merchants, 1 and this system of official deliberation was 
applied to all departments of the central government. 
Every matter was to be maturely discussed before being 
remitted for the pasha’s orders. Then in 1829 the 
principle was more widely extended. Ibrahim Pasha 
presided over a gathering of 400 persons specially 
convened, comprising the chief civil and military 
officers, the mudirs, and even a number of village 
shaikhs, to discuss the best way of correcting abuses and 
improving the condition of the peasants. It sat every 
night for some time, and the members were sworn to 
secrecy. 2 In 1832 a similar plan was tried in Syria. 
A small body of Grand Notables, twenty-two in 
number, was convoked to deliberate on the affairs of 
the people. 3 In 1 834 the shaikh of the A 1 Azhar mosque 
and the shaikh of the merchants’ guild were directed to 
nominate suitable ulema and merchants to sit in the 
Superior Council, and at the same time the mudirs 
were ordered to convene in each mudirliq an assembly 
including cultivators, the village shaikhs, and others, to 
choose two village shaikhs to represent the cultivators 
of the mudirliq in the council. 4 These things were 
generally misconceived and misrepresented by the 
tourists whose information gave the tone to European 
opinion. On the one side was the young Disraeli repre- 
senting the pasha as saying he would have as many 

1 To Kotkhuda Bey, Shaban 12, 1235 (Abdine Archives). 

2 Barker to Gordon, September 22, 1829 (F.O. 78-184). 

3 Ibrahim’s proclamation, Safar 15, 1248 (Abdine Archives). 

4 To Habib Effendi, Rabi-al-awal 25, 1250, and a circular to 
the mudirs of the same date {ibid.). 


parliaments as William IV but would take care to 
choose them himself. On the other were Philosophic 
Radicals and Saint-Simoniens who claimed Muham- 
mad ’Ali as a convert to western democracy. The one 
suggests that the pasha’s experiments were mere tricks 
designed to impress European opinion; the other that 
they meant the establishment of representative govern- 
ment . 1 Of course they were neither one nor the other. 
Ordinary public business in the East had always been 
decided by a group of officials — a divan or durbar — 
presided over by the pasha in person, or some superior 
official, and sitting virtually in public, with a constant 
succession of petitioners and spectators. As Bartle 
Frere once said, important as is a knowledge of public 
opinion in a western country, its importance is even 
greater in the East. The oriental ruler needed urgently to 
know what men were saying in the bazaars and caravan- 
sarais. In part he could depend upon his spies — the 
most permanent of all instruments of government in 
Asia. But another side perhaps could be revealed by 
the assemblies which Muhammad ’Ali from time to 
time convoked. He was much too shrewd to dream of 
borrowing wholesale from western practice. But he was 
also much too shrewd not to see that a discreet borrow- 
ing of western ideas, adapted in such a way as not to 
disturb accustomed forms, might prove very beneficial 
to his government. He must have been influenced too 
by another motive. A man with so strong and clear a 
sense of the value of knowledge can hardly have been 
unaware that his tentative assemblies were instruments 
not only of government but also of political education. 
Had chance but provided Egypt with an heir to 
Muhammad ’Ali’s talents as well as to his dominions, 

1 Monypenny and Buckle, Disraeli, x, 1 76-7. St John, Egypt and 
Mohamed Ali, 11, 472 sqq. Cf. Bentham’s memorandum (Brit. 
Mus. Add. MSS 25663, ff. 139 sqq.). 



the country would have afforded to western nations an 
example of political reform as remarkable as that of 
Japan. But a single life, largely spent in building up a 
political dominion, cannot conceivably do more than 
plan the outline of institutional development. 

His financial management was extraordinarily suc- 
cessful. Onlookers were always anticipating his finan- 
cial ruin and declaring that his wars and internal 
projects would ruin him and the country alike. In 
1827, for instance, when he was burdened with the cost 
of the war in the Morea at the moment that his re- 
sources were straitened by two successive failures of the 
Nile to attain its usual height, he still went on building 
factories and constructing a mole and dockyard at 
Alexandria. 1 Four years later he was contemplating 
projects ten times as great. 2 He kept out of the hands 
of European moneylenders. 3 In 1837 it was thought 
that the fall in the price of cotton (of which he exercised 
a monopoly) would hit him hard : and yet he managed 
to pay off the arrears due to his troops. 4 Barker thought 
he really must have found Aladdin’s lamp. 

But his magic lay merely in prudence and attention. 
The public accounts, when he obtained the govern- 
ment, were kept by Copts, who made them a perfect 
model of intricacy. This had two advantages — it made 
their services indispensable and it hid their defalca- 
tions. The public accounts were not centralised. 
Various taxes would be assigned to various services 
according to the approved Turkish mode. 5 There was 
no budget nor any possibility of one. Here as elsewhere 
the pasha was willing to learn and borrow from the 

1 Barker, Syria and Egypt , 11, 60-1. 

2 Idem, p. 158. 

3 Cf. Campbell, October 12, 1833 (F.O. 78-228). 

4 The same. May 25 and July 13, 1837 (F.O. 78-319 and 320). 

5 E.g. the assignment of the Damascus Miri for the haj. See 
p. 42 supra. 


European. He directed Boghoz Bey, the Armenian, 
the most trusted and the most trustworthy of all his 
servants, to obtain a scheme of accounts as used in 
public offices in Europe; 1 and the Frenchman Jomard 
was employed to frame a new system. 2 But that still 
left untouched the vicious method of apportioning 
“different districts. . .to different ministers to provide 
for their expenses instead of sending their revenue to a 
common treasury. The present state of things leads to 
great abuses, as every minister has a treasury of his 
own, and seven doors are opened (those are the dif- 
ferent ministries) to fraud and abuse, when one in a 
country like this is already but too many”. 3 When 
Bowring visited Egypt in 1838 the pasha sought his 
advice in the matter of accountancy. All the public 
accounts were produced for inspection, and Bowring 
made a number of recommendations for their im- 
provement. First came the introduction of a budget of 
receipts and expenditure at the beginning of each year ; 
then the payment of all revenues into the central 
treasury; the complete separation of the power to 
receive and issue public money; the establishment of 
the power in the Finance Minister to sanction or reject 
proposed expenditure; and finally provision for the 
prompt payment, balancing and audit of all public 
accounts. 4 

The fragmentary evidence as yet available does not 
permit of any full or exact narrative of the pasha’s 
financial history. But he seems usually to have suc- 
ceeded in keeping his expenditure well below his 
revenue. For instance in 1820-1 (the Coptic year used 
in the Egyptian accounts ended on September 28) 6 the 

1 To Boghoz Bey, Rabi-al-awal 22, 1249 (Abdirie Archives). 

2 Jomard, Coup d’CEil, p. 23. 

3 Campbell, February 26, 1838 (F.O. 78-342). 

4 Bowring, Report (Pari. Papers , 1840, xxi, 62). 

6 Murray, March 17, 1848 (F.O. 78-757). 



revenues amounted to 240,000 purses against an ex- 
penditure of under 190,000. In 1832-3 they were a 
little over 500,000 against an expenditure of 415,000. 
In 1846-7 they were over 600,000 against an expendi- 
ture of 460,000. It is obvious that there must have 
been years of heavy expenditure when the accumulated 
surplus was much reduced ; but surpluses seem to have 
been more frequent than deficits. Land revenue — miri 
— was naturally much the largest item among the 
receipts, but it scarcely amounted to more than 50 per 
cent., while on the expenditure side the army and 
marine seem to have absorbed about the same pro- 
portion of the total. 1 

Egyptian land tenures at the beginning of the nine- 
teenth century were in much the same confused con- 
dition as when the East India Company undertook the 
administration of its Indian provinces. All the four 
orthodox schools of Muslim law recognised it as con- 
quered territory with all its lands vesting in the 
Caliph. In token of this the leader of the Friday prayers 
at every mosque throughout the country mounted the 
pulpit holding in his hands a real or imitation sword. 
But, as everywhere else, the ruler had always been 
alienating his demesnes by grants, sometimes revocable 
at will, sometimes purporting to be irrevocable. The 
actual form probably made small difference, and 
Muslim lawyers held the convenient doctrine that even 
the most binding grant could be recalled in the interest 
of the state — a matter of which the ruler himself would 
be sole judge. 

However, the confusion of the seventeenth and 
eighteenth centuries bred a host of proprietors, the 
most important of whom were the Mameluke chiefs 
and the tax-farmers called multazim. The lands held by 

1 Bowring, ut supra, p. 44. Murray, ut supra. St John, op. cit. n, 
469 sqq. 




the former of course paid no revenue, while those in the 
possession of the latter included ever-growing areas 
held tax-free — ussieh lands — in consideration of the 
labour and expense of collecting the revenue on the 
remainder. One of the pasha’s earliest actions, as I 
have already stated, 1 was to appropriate the Mame- 
lukes’ lands, and to hold an enquiry into the tenures by 
which the other lands were held. Between 1808 and 
1814 he appropriated the whole, giving the multazim 
pensions instead of the ussieh lands held by them. 2 In 
this the pasha does not seem to have exceeded the 
theoretical limit of his legal rights, although it must be 
remembered that “legal rights” did not carry the same 
connotation as in Europe. The measure was extreme, 
and Muhammad ’Ali himself would perhaps only have 
justified it by his extreme financial need at this period. 
He could not establish a stable government unless he 
resumed that large proportion of the land, amounting 
to nearly three-quarters, which had fallen into private 
hands by the corrupt negligence of his predecessors. 
The doctrine of necessity may be called in to justify 
anything. But the measure did not affect the fellah at 
all, but only a small class of proprietors, and the pasha’s 
English critics might with propriety have remembered 
that Lord Cornwallis had expropriated not a small class 
of landlords but a large class of peasants in Bengal. No 
unjust act can be defended. But the guilt of injustice to 
the few seems less than the guilt of injustice to the many, 
for it inflicts a lower amount of avoidable suffering. 
Both the governor-general and the pasha undoubtedly 
believed their respective policies to be in the interest of 
the country at large. 

1 Vide p. 32,, supra. 

2 Artin Bey, Propriety fonciere en £gypte , pp. 84-6. St John, op. cit. 
11, 456 sqq. Jomard, Coup d’CEil, p. 11. Missett, March 22, 1814 
(F.O. 24-5). Cf. Young, Corps de droit Ottoman , vi, 45 sqq. 



This general resumption of the land was followed up 
by a cadastral survey of the country. To this the pasha 
paid a close attention. The mudirliq registers are said 
almost everywhere to bear the seal of his inspection. 1 
But in this matter his work was impaired by the 
quality of his servants. His surveyors were unskilled, 
the supervisors dishonest. 2 In fact he had to encounter 
all the difficulties which in Bengal prevented the East 
India Company from ever undertaking a revenue 
survey, and in the other Indian provinces produced a 
multitude of early errors. The pasha found, for instance, 
that rich occupiers bribed the surveyors to show their 
lands as uncultivated and not irrigated, while the re- 
ductions thus made in the demand were recovered by 
over-assessing the lands tilled by poorer cultivators. 3 
But though defective in detail and needing much re- 
vision, which was only applied gradually as the defects 
came to notice, the survey revealed much cultivated 
land which had till then escaped assessment by de- 
liberate fraud. 4 

Another most important measure was the promotion 
of irrigation. The pasha claimed to have introduced no 
less than 38,000 sakias — water-wheels — or more than 
half of those in operation in 1844. 5 He cleared the old 
irrigation canals and dug new ones, and in Upper 
Egypt especially brought a great amount of land under 
tillage. Campbell mentions a new canal designed to 
irrigate a million acres. 6 Bowring found that 100,000 
feddans had in fact been brought into bearing. 7 The 

1 Artin Bey, op. cit. p. 88. 

2 Campbell, February 26, 1838 (F.O. 78-342). 

3 To the Mudir of Ghiza, Safar 8, 1250 (Abdine Archives). 

4 Jomard, loc. cit. 

8 Bowring, Report (Pari. Papers, 1840, xxi, 12). Barnett, 
December 12, 1844 (F.O. 78-583). 

® Campbell, January 1, 1834 (F.O. 97-411). 

7 Bowring, loc. cit. 



man who was of most service to the pasha in these 
matters was the French engineer Linant, who also 
projected a great work which was to have controlled 
and extended the irrigation of the Delta. This was the 
famous Nile barrage, to be built near the fork of the 
Delta below Cairo. This, it was hoped, would permit 
the complete irrigation of the Delta in even the worst 
of Niles, and bring into cultivation between two and 
three hundred thousand acres above the barrage. 1 The 
difficulty was mainly technical. Linant had had no 
previous experience of such works. The project was 
long discussed, and at last a plan was prepared for sub- 
mission to the French Board of Civil Engineering. 2 
Most people doubted the possibility of carrying through 
so great a work, which was expected to take five years 
and cost at least a million and a half sterling. 3 It was 
not till 1847 that the foundation stone was actually 
laid, and then when accomplished the work failed to do 
what was expected of it. The great river found its way 
under the inefficient foundations. The pessimists . 
proved for the moment to have been right. Modern 
engineers distribute the blame between Muhammad 
’Ali’s impatience and Linant’s inexperience. Indeed 
the whole episode illustrates alike the pasha’s strength 
and weakness — on the one side his vision and zeal for 
betterment, on the other his uninstructed haste and the 
imperfect instruments at his command. 

Despite this failure, the cultivated area was con- 
siderably enlarged under the pasha’s rule; and grants 
were freely made in order to promote agriculture. 
From 1829 onwards waste lands were ceded to indivi- 
duals on condition of cultivation. At first the grants 
gave only a heritable usufruct ; but in 1 842 they were 

1 Campbell, November 1, 1834 (F.O. 78-247). 

2 Barnett, October 20, 1842 (F.O. 78-502). 

s Bowring, loc. cit. 



transformed into grants of absolute property. About 
the same period the lands rendered cultivable by the 
pasha’s new irrigation works were granted out — under 
the title of chifliqs — on condition of progressive tillage. 
Much of the new area was bestowed on members of the 
pasha’s own family. 1 So private property in land 
revived. At the same time it extended imperceptibly 
again over the rest of the country. The individuals 
registered in the cadastral survey gradually became in 
fact proprietors. As in India land came again to have a 
sale value, and Bowring could hear of no one who had 
been expropriated in recent times except for failure to 
pay the land revenue, 2 just as would have been done in 
similar circumstances in India. 

The land revenue was payable either in kind or in 
money. Certain areas, specially suitable for the growth 
of crops such as cotton or indigo, in which the pasha 
had established a monopoly, were required to deliver 
specified quantities of the appropriate article. Else- 
where the occupier might cultivate what he pleased, 
subject to the payment of an assessment computed on 
the quality of the soil and the value of the crop that 
might best be grown on it. Down to 1834 the assess- 
ment was levied on a holding, whether cultivated or 
not, provided there was water enough even partially to 
irrigate it. But in that year the pasha adopted the much 
more equitable plan of levying the revenue only on the 
land which could be fully irrigated. 3 Another im- 
portant reform was the abolition of the age-old custom 
of making good the failure of payment on one holding 
by levying an extra rate upon the rest. The practice 

1 Artin Bey, op. cit. pp. 95 sqq. Cf. to the Director of Roznamah, 
Zilhaj 24, 1256 (Abdine Archives); Barnett, January 15 and 
December 12, 1844 (F.O. 78-582 and 583). 

2 Bowring, Report (Pari. Papers, 1840, xxi, 123-4). 

3 Campbell, April 27, 1834 (F.O. 78-245). 


seems to have been almost universal in the East. It had 
been as familiar in India as it had been in Egypt. It 
was defended on the score that it helped to prevent 
village shaikhs and other influential persons from 
thrusting an unfair proportion of the assessment on the 
smaller holders . 1 

The actual amount and rate of the assessments seem 
to have been considerably increased. It is even said 
that the actual money assessment was roughly doubled . 2 
But this statement by itself is certainly misleading, for 
it leaves out of account various extra taxes, some recog- 
nised, others concealed, which the officials had col- 
lected and which had been not only prohibited but in 
some part suppressed. Nor can the extreme reluctance 
of the fellah to pay be taken too seriously into account. 
Centuries of hard experience had taught him, as it had 
taught the Indian ryot, that willing payment was in- 
expedient. It always was interpreted as indicating a 
superfluity of money, and therefore led to a demand for 
more. The feeling had been strengthened by the period • 
of loose, weak government which had preceded the 
pasha’s as it had the East India Company’s Govern- 
ment. French observers of the time of Napoleon bear 
witness to the extraordinary difficulty which the Mame- 
lukes had found in getting in the revenue. “The 
peasants . . . only pay at the last extremity, and sou by 
sou ; they hide their money ; they bury their goods and 
chattels . . . When they see a body of troops coming, 
they flee with their wives, children and cattle, leaving 
only their empty huts behind. If they think themselves 
strong enough to resist, they fight, calling in to their aid 
neighbouring villages and even the Bedouin.” The 
Mamelukes had to maintain troops in each of their 

1 Bowring, Report (Pari. Papers, 1840, xxi, 15). Puckler- 
Muskau, op. cit. I, 23-4. 

2 Cf. St John, op. cit. n, 450 sqq. 



provinces solely engaged in trying to force the villages 
to pay and often enough not succeeding . 1 Muhammad 
’Ali’s power was more firmly established than this. 
Open resistance seems seldom to have been attempted. 
But passive resistance still continued. It was apparently 
a point of honour with the fellah to pay nothing till he 
had received a certain number of stripes, and the one 
who resisted the longest was the most esteemed . 2 

This was not the only parallel that may be traced 
between the Egyptian fellah and the Indian ryot. By 
ancient usage both were regarded by the governments 
which Providence was pleased to establish over them as 
existing to fulfil one function in life — the tillage of the 
soil. The duty of the cultivator was to cultivate. If he 
neglected that duty, the ruler must punish him with 
sharp penalties. As a recent writer has well said of 
Hindu and Muslim India, “the agrarian system was a 
matter of duties rather than rights ”. 3 Both Muham- 
mad ’Ali and the East India Company inherited that 
conception unchanged from the past. As was natural, 
the pasha clung to it more firmly than did the Company’s 
servants. He was not willing to permit land to lie un- 
tilled. When he learnt, for instance, that the lands 
granted to village shaikhs in return for their services 
were lying unwatered and covered with weeds, he 
directed the guilty shaikhs to be beaten at the corner of 
their fields, as an example to others . 4 He judged a 
state of tutelage to be necessary for the fellah’s own 
well-being. He was extremely active in enquiring into 
any grievances, and under him, Salt observed, the 
peasants “are in general better treated and more 

1 Poussielgue’s letter, 1 Vendemiaire, l’an vn, ap. Intercepted 
Letters, pp. 46 sqq. 

2 Bowring, Report (Pari. Papers, 1840, xxi, 47). 

3 Moreland, Agrarian System of Moslem India, p. xi. 

4 To the mudirs of Lower Egypt, Rabi-al-awal 2, 1252 (Abdine 


content than for many years past”. 1 Salt’s evidence, it 
may be noted, is of more than usual value, because his 
frequent journeys in search of antiquities brought him 
into close and varied contact with the peasants of the 
country. Other observers agreed with the pasha’s 
theory of the need of control. “From my own ex- 
perience of the Arab character such as it exists at 
present”, Thurburn wrote, “I must confess that there 
is some truth in the opinion that the Egyptian peasant, 
if left to the free exercise of his own will, would confine 
himself to the wants of the moment and long continue 
to direct his attention to the cultivation of those 
articles only which are produced with the least outlay 
of labour and capital. . . .” 2 

However the condition of the fellahin fell away at a 
later date. This was probably due less to the weight of 
the land revenue demand in itself than to the fact that 
the system of conscription, to which I shall return later, 
reduced the productive capacity of the villages while 
the revenue demand was maintained at the former rate. 
From 1829 begins a series of complaints of peasants 
abandoning their villages, and the severest orders were 
issued against both the emigrant peasants and the 
officials in whose jurisdiction they should be dis- 
covered. 3 Muhammad ’Ali himself ascribed this 
practice to two causes — ill-treatment by the local 
officials and ignorance. “There are two sovereigns”, he 
said, “the Sultan Mahmud and the fellah,. . .and the 
fellah must not be regarded with an evil eye.” 4 
Peasants, he says again, are not to be imprisoned for 
neglect of cultivation, for the first duty of government is 

1 Salt, April 28, 1817 (F.O. 78-89). 

2 Thurburn, ap. Bowring, Report (Pari. Papers, 1840, xxi, 64). 

8 To Zeki Effendi, Shawal 14, 1244; Circulars, Muharram 13, 

1259, and Muharram 17, 1260 (Abdine Archives). 

4 To the Inspector-General of Factories, Jumadi-us-sani, 1 252 




to secure the well-being of the people. 1 Villagers were 
to be allowed to appeal to the mudirs if they were ill- 
treated, and to petition the pasha himself if they failed 
to obtain justice. 2 

With this growing restlessness of the population went 
an accumulation of uncollected arrears of revenue. In 
1833 the mudirs were warned that they would be held 
personally responsible for these balances. 3 In 1835 the 
pasha made a tour especially to enquire into this 
question, 4 and was led to make considerable remis- 
sions. 6 Finally he adopted the dubious device of com- 
pelling his principal officers to take over the villages 
most heavily in debt with the obligation of paying off 
gradually the arrears as well as meeting the current 
revenue demand. They were very reluctant to under- 
take the task, but were told that they had grown rich in 
the pasha’s service and could not be allowed now to 
desert him. 6 In general, the revenue management 
suffered from much the same defects as the Company’s 
early revenue management in India. The demand was 
pitched too high to be met in an average year, the 
revenue subordinates were careless and corrupt, and 
the assessments were unequal, so that some villages 
could, whilst others could not, pay the demands 
made upon them. In other words, the pasha did not 
succeed in ridding the revenue system of its traditional 

The land revenue system, while of prime and essential 
interest to the country as a whole, was not of general 
concern to foreign countries. But Muhammad ’Ali’s 

1 Circular, Rajab 1 , 1 252 (Abdine Archives) . 

2 To the Divan al Chora, Rabi-al-awal 17, 1260 {ibid..). 

8 To the mudirs, Safar 10, 1249 (ibid.). 

4 To the mudirs, Zilkaidah 17, 1250 (ibid.). 

8 Campbell, September 15, 1835 (F.O. 78-258). 

6 Artin Bey, op. cit. pp. 128 sqq. Barnett, April 16, 1845 (F.O. 


trade policy closely affected both and received a great 
deal of attention. Under Mameluke rule the Turkish 
Capitulations had scarcely been reckoned as applying 
to Egypt. Life had been too insecure, trade too irregu- 
lar, the beys too rebellious, and the European trade 
with Egypt too insignificant for France or England to 
attempt to stand upon their theoretical rights. This 
attitude had continued long after the pasha had estab- 
lished himself at Cairo, and for years no one thought 
of protesting formally, however much he might com- 
plain in private, of the regulations introduced in 
trading matters. It was not till the ’thirties of the 
century that difficulties began to arise. Then Campbell 
criticises the apathy and neglect of the earlier consuls, 
“many of whom, being in trade and in debt to the 
pasha, were afraid to assert the just rights of their 
countrymen”; 1 while Mole, writing to de Lesseps two 
years later, lamented the early tolerance of the Euro- 
pean representatives which had complicated matters 
and rendered complaint difficult. 2 In its origin and » 
early stages the pasha’s trade policy had been inspired 
by his need of finding money and by those evident 
advantages of monopoly which have always appealed 
to eastern rulers no less than to western merchants. 
Salt in 1820 and again in 1826 sent home sharp com- 
plaints of the mercantile evils flowing from Muham- 
mad ’Ali’s position as the chief merchant of the country 
which he ruled. 3 He not only compelled the fellah to 
cultivate, but in some areas he determined what crops 
should be grown, and required the produce to be de- 
livered into the government warehouses at a fixed rate. 4 

1 Campbell, July 15, 1835 (F.O. 78-258). 

2 M0I6 to de Lesseps, March 3, 1837 (F.O. 78-319). 

s Salt, June 30, 1820 and April 4, 1826 (F.O. 78-96 and 147). 

* The student of colonial history will be reminded of the Dutch 
system of “cultures” in Java. 


The evils of such a practice are too evident to re- 
quire statement. But there was another aspect of the 
matter. The resources of the country were being de- 
veloped as they had not been for ages. “ It is also to be 
remembered”, says Salt, “that the pasha had in some 
sort created all these fine articles of produce which now 
form the most valuable commodities for export, as 
cotton, indigo, sugar, by the judicious application of 
large capital to these several branches of industry, 
which the peasants would neither have had the means 
nor spirit to have improved .” 1 Later on the poppy 
culture was introduced on a large scale in Upper 
Egypt. Plantations of mulberry trees were made. 
Factories were set up for the manufacture of sugar and 
distillation of rum. A tannery was established at 
Rosetta to supply belts, boots and saddles for the 
army . 2 Mills were built for the manufacture of cotton 
cloth. At one time the pasha had almost realised at 
all events one aspect of the socialist ideal. 

Many of these activities were based on false notions. 
The more elaborate factories were a failure. The 
machinery was neglected, the running parts left un- 
oiled, the management ignorant and careless. Oxen 
provided the driving power, whereas of course the 
Nile itself should have been harnessed to the work. The 
fellahin detested the unaccustomed regularity of toil 
and had to be gathered as the army conscripts were, by 
force. The pasha, Bowring observed with complete 
truth, “takes hands from the fields, where they would 
be creating wealth, to employ them in . . . fabrics 3 where 
they are wasting it ”. 4 He was said to have sunk twelve 
million sterling in these factories and the machinery 

1 Salt, May 20, 1825 (F.O. 78—1 35). 

2 Barker, Syria and Egypt, n, 157-8. 

3 I presume “fabriques”. 

4 Bowring to Campbell, December 17, 1837 (F.O. 78-342). 


with which they were equipped, 1 and all to no purpose. 
But wasted as much of this endeavour was, it is worth 
noting with respect because it marks a modification of 
the pasha’s conceptions of his duty. He began by 
seeking only to raise money. He ended by seeking, 
however mistakenly, to develop and civilise the country. 
In this matter he was carried away into an excessive 
and unwise imitation of the West; but he had come to 
be something far nobler than the greedy adventurer 
seeking nothing but his own power and wealth. Even 
his monopolies had their good side. He may have 
squeezed the fellah, but he squeezed him less closely 
than the foreign merchants would have done had they 
been left free to buy and sell as they pleased, and the 
burden of mercantile advances would have been yet 
heavier than the arrears of the pasha’s revenue. Of this 
Muhammad ’Ali was convinced. 2 

This policy however could not but provoke the 
irritation of the British government. Since Egypt was 
part of the Ottoman Empire it must be bound by the 
Turkish Capitulations, and these, as read by British 
merchants, signified the right of free and unrestricted 
trade. “They may bring”, ran the 53rd article, “and 
may in like manner buy and export all sorts of merchan- 
dise without anyone presuming to prohibit or molest 
them.” But in the first place this apparent right to 
liberty of trade was limited by an ominously obscure 
phrase elsewhere excluding “prohibited articles” from 
its operation. Salt argued correctly that “it leaves 
almost everything to the caprice of governors and 
commandants, who may take advantage thereof to 
introduce any articles they may choose on that list” — a 
view which Stratford Canning, in an appended note, 

1 Bowring, Report {Pari. Papers, 1840, xxi, 31 sqq.). Campbell 
to Bowring, January 18, 1838 (F.O. 78-342). 

2 Campbell, March 24, 1839 (F.O. 78-373). 



described as “well-grounded and judicious”. 1 Hence 
the negotiations which Palmerston undertook through 
Ponsonby for a revision of British trading regulations 
in the Turkish empire, leading up to the commercial 
convention which was signed in 1838. It contained 
clear stipulations for the abolition of monopolies, which 
Palmerston was resolved should be duly enforced in 
Egypt, and which he argued would benefit that 
country as much as those who traded with it. “ It must 
be evident”, he wrote, “to every person who is at all 
conversant with the principles which regulate the 
wealth of nations that [the pasha’s] system . . . tends to 
keep Egypt and Syria in a state of abject poverty.” 2 As 
soon as the convention had been concluded, Ponsonby 
called on the Syrian consuls for an account of the mono- 
polies established by the pasha’s government. The 
consuls at Aleppo and Damascus reported there were 
none. The consul at Beyrout sent in a long list which 
showed on examination that he confused monopolies 
with excise duties. 3 In Egypt the position was clearer. 
The pasha exercised monopolies. But mainly owing to 
delays in the transmission of the necessary farmans 
from Constantinople, the question was not really 
taken up till after the crisis of 1840 had been settled. It 
then proved difficult, because the great quantities of 
cotton, sugar and other produce controlled by the 
pasha were delivered to him either as the owner of 
chifliqs 4 5 or as part of the land revenue of the country. 
Palmerston drafted angry and threatening despatches. 6 
But commercial opinion at Alexandria and Cairo had 
been too much annoyed by his late policy and perhaps 

1 Salt, May 20, 1825 (F.O. 78-135). 

2 Memorandum, September 13, 1838 (F.O. 96-19). 

3 Campbell, March 30, 1839 (F.O. 78-376). 

4 Cf. p. 213, supra. 

5 To Barnett, August 26, 1841 (F.O. 78-451). 


was insufficiently informed of “the principles which 
regulate the wealth of nations”, to aid the consul with 
complaints. 1 There was moreover another reason why 
they regarded the commercial convention with marked 
distaste and were unwilling to see it put into operation 
in Egypt. So far as that country was concerned, it had 
been prepared, I think, rather to reduce the pasha’s 
revenues by getting rid of his monopolies than in order 
to benefit British trade. While it was decidedly advan- 
tageous at Constantinople, Smyrna, and other ports 
under the Sultan, in Egypt the British exporter of 
cotton would be required to pay 1 2 per cent, instead of 
3, and in Syria all produce, when exported by British 
merchants, would pay 12 per cent, instead of 2. 
Foreign merchants would of course continue to pay 
only the old rates, so perhaps there was good ground 
for the mercantile dislike of Palmerston’s policy. 2 Nor 
was this all. The rates of the convention were specific, 
not ad valorem rates. When they came to be actually 
introduced in 1841 they proved to be 22 per cent, ad 
valorem on cotton, 20-25 P er cent, on wool, well over 
12 per cent, on grain, while import dues, which had 
been meant to amount to 5 per cent., worked out at 
nearer 9 per cent. The upshot of the matter was that 
the pasha finally agreed to levy ad valorem rates of 12 
per cent, on exports and 5 per cent, on imports payable 
in Egyptian currency, 3 while in regard to the monopo- 
lies he promised in future to sell his produce by public 
auction. 4 It is difficult to reflect on these blundering, 
disingenuous negotiations with any feeling of national 

One essential condition of the maintenance of the 

1 Barnett to Stratford Canning, December 1, 1 84 1 (F. 0 . 78-451). 

2 Campbell, September 3, 1839 (F.O. 78-376). 

8 Barnett, May 20 and 26, 1842 (F.O. 78-502). 

* The same. May 15, 1842 (ibid.). 



pasha’s position had been the assemblage of forces 
capable of opposing those of his sovereign, the Sultan. 
That he should create as large an army as he could was 
a thing of course ; but that he should seek to create a 
navy marks the vigour and the limitations of his mind. 
It was a prime need if he was ever, as he hoped, to 
dominate the empire. But it had to be built up from 
the very foundations, in a country with no trace of 
naval traditions, by a ruler with not an atom of 
technical knowledge. He began by building abroad — 
at Bombay, at Leghorn and Marseilles. In 1821 he 
asked both the English and the French governments to 
build him frigates. 1 A little later he made a dock at 
Alexandria and began to build on his own account, 
employing French shipwrights to control the work. In 
1828 he began to build a naval arsenal for the supply 
and maintenance of his naval force. He set to work at 
once to replace the fleet that had perished at Navarino, 
confident now that his vessels would be superior to any 
that the Sultan could assemble. Instead of building 
‘frigates he began to build ships of the line, mounting a 
hundred guns or more. 2 In 1829 Cerisy, from the royal 
dockyard at Toulon, was placed in charge of the 
Alexandria docks. In 1831 was launched the first of his 
100-gun ships, named after himself. 3 In 1833 he had 
six ships of the line, ranging from 84 to no guns, 
besides seven frigates, and in 1837 he had eight of the 
former besides one more under construction. 4 In the 
arsenal was a staff of over 3000 hands under the direc- 
tion of sixty Europeans. Attached to it was the naval 
school of Ras-al-tin with 1200 cadets. 

1 Sal t, November 6, 1821 (F.O. 78-103), 

2 Barker to Gordon, May 26, 1829 (F.O. 78-184). 

3 The same to Sir P. Malcolm, January 15, 1831 (F.O. 78-202). 

4 Campbell, April 24, 1833 and July 14. 1837 (F.O. 78-227 and 


This development was pressed on by the pasha with 
strong enthusiasm and the closest personal attention, 
driving on his subjects to co-operate however reluct- 
antly with him. He would often amuse himself by 
cruising off Alexandria, and I have already mentioned 
his hurrying off to chase the Greeks with a single 
vessel. 1 He established a code of regulations, based on 
those in force in the British and French navies, but 
carefully adapted to Turkish usage. 2 But while his 
vigour could get ships, and good ones, built, not even 
he could raise sailors where no mercantile marine had 
ever existed. “A hot press is on foot”, our consul- 
general writes in 1832, “not of sea-faring men, for none 
such are to be found, but of every individual without 
discrimination.” Within forty-eight hours some 1000 
men were gathered up in Alexandria to complete the 
complements. 3 With a stiffening of real sailors and 
under the command of experienced and skilful officers, 
something might have been made even of such hap- 
hazard crews. But, as the same writer says, “they have 
no experienced and patriotic officers nor even ordinary 
seamen”. 4 In 1831 Muhammad ’Ali hoped to make 
good these deficiencies by enlisting English officers 
and men, and employed Colonel Light — a son, I think, 
of the grantee of Penang — to engage them for him. He 
wanted two post-captains, two commanders, several 
lieutenants, and forty or fifty petty officers and able 
seamen. 5 But nothing for the moment was done. Not 
until 1834 did the British government decide to permit 
naval officers on half pay to serve under the pasha, 6 

1 Cf. p. 74, supra. 

2 Barker to Sir P. Malcolm, ut supra. 

3 The same to Stratford Canning, February 20, 1832 (F.O. 78— 


4 The same to Mandeville, January 2, 1832 (ibid.). 

5 The same, August ix, 1831 (F.O. 78-202). 

6 Campbell, October 25, 1834 (F.O. 78-247). 



and by then the pasha had already employed a number 
of French officers. Besson Bey was vice-admiral with a 
European-educated Turk, Hasan Bey, under him as 
rear-admiral. By way of showing his personal interest 
and placing his navy under powerful patronage in the 
future, the pasha resolved that one of his sons, Said 
Bey, should be brought up to the sea. He was therefore 
sent on ship-board, nominally as a midshipman, at the 
age of thirteen. He was provided with a tutor, and a 
special French officer was detailed to teach him his 
professional duties. After five years’ service he was 
given command of a corvette. But he afflicted his 
father by his inactivity and premature fatness. He was 
periodically weighed, and whenever an increase was 
recorded he would receive a letter exhorting him to 
discern good from evil, to ensue manly qualities, and 
to rid himself of fleshliness hateful in the eye of all . 1 

The pasha’s navy, like his factories, lacked a solid 
foundation. It could only be maintained by the closest 
attention of its founder. It appealed in no way to any 
class of the people. It had no natural recruiting ground. 
It was more unpopular than the army. It lay in 
Alexandria during the brief Syrian war ; the Capitan 
Pasha’s surrender robbed it of its chance of performing 
the service for which it had been called into being; and 
soon after the pasha’s death all the serviceable vessels 
were sold off to the Porte. The experiment had proved 
a failure. 

The pasha’s military labours were much more fruit- 
ful and important. I have already mentioned how his 
original mercenary and foreign army had been trans- 
formed into one following European discipline, formed 
on the European mode of organisation, and raised in 

1 Campbell, August 19, 1834; October 7, 1836; and May 14, 
1839 (F.O. 78-246, 284, 373). Also the Pasha to Said Bey, 
Ramzan 9, 1 253 (Abdine Archives) . 


the country by conscription. 1 By 1832 a very con- 
siderable disciplined force had been raised. It then 
consisted of twenty regiments of infantry, ten of 
cavalry, with a small body of Turkish and a much 
larger one of Bedouin irregulars, totalling 83,000 men. 2 
Three years later the strength had been raised in Syria 
alone to 69,000 — an increase of 50 per cent. 3 The pasha 
probably had at this time over 100,000 men under 
arms. At first they were equipped with muskets im- 
ported from France and England: but, the quality 
being poor, the pasha set up his own factory, and was 
provided with 2000 stand of arms of the best and latest 
pattern from the Tower as samples. 4 5 The training of 
the troops was at first under the superintendence and 
control of French and other continental officers, such 
as Colonel Seve. The officers, for whom special schools 
were established at Ghiza and elsewhere, were drawn 
almost exclusively from Turkish and other foreign 
families, 6 many having been the personal slaves of the 
pasha who had chosen them as showing fitness for 
military service. The privates however were exclusively 
Egyptians with a considerable number of Syrians so 
long as Syria- if rfi'aihed in Muhammad ’Ali’s possession. 

Tbe methods by which they were raised have 
^•inerally been considered the worst blot upon his 
administration. A census had been attempted but 
abandoned, owing to the universal opposition in which 
even the pasha’s officials had taken part.® The only 
possible procedure was to call upon the mudirs to 

1 Vide p. 65, supra. 

8 Barker, July 21, 1832 (F.O. 78-214). 

3 Campbell, December 12, 1835 (F.O. 78-258). 

4 The same, December iq, 1833, and to Campbell, September 

16, 1834 (F.O. 78-228, 244). _ 

5 Barker seems to have been misinformed when he says they 
were all Arabs. Despatch of June 23, 1829 (F.O. 78-184). 

6 Campbell, Report (F.O. 78-408 b). 



supply a fixed quota of men. This quota was divided 
out among the villages. The shaikhs seized as many as 
they could, released those who offered the largest 
bribes, and sent in the remainder chained two and two 
like felons . 1 When the pasha’s power was at its height 
he was perhaps claiming one man out of every six for 
military service. 

Nothing in the whole government was so feared and 
hated as the conscription. The stories of travellers who 
assert that many cut off their right forefinger in order 
to escape service in the army 2 might be suspected of 
philanthropic exaggeration. Campbell, relating that 
they cut off a finger, drew their teeth, and blinded 
themselves , 3 might have been misinformed. But such 
statements are borne out beyond the possibility of error 
by the pasha’s own correspondence. Men who put rat’s 
bane in their eyes, he writes, are beasts in human form 
who must be sent to forced labour for life . 4 * If the 
barber’s wife who assisted them is guilty, she is to be 
executed and her body exposed for three days . 6 
Another woman for a like offence is to be thrown into 
the Nile . 6 Conscripts are to be warned that if they 
maim themselves, not only will they be sent to forced 
labour for life but also another member of their family 
will be seized in their stead . 7 The officials are warned 
that the practice must be due to their neglect and that 
if it continues they also will receive the same punish- 
ment . 8 Forced labour proving no deterrent, death was 

1 Barker to Sir P. Malcolm, July 8, 1829, and Murray, June 1, 
1848 (F.O. 78-184, 757). 

2 St John, op. cit. n, 175. 

3 Campbell, February 26, 1838 (F.O. 78-342). 

4 To the Kotkhuda Bey, Shaban 17, 1245 (Abdine Archives). 

6 To the Mamur of Fayum, Ramzan 1, 1245 (ibid.). 

6 To the Mamur of Tanta, Zilkaidah 13, 1245 (ibid.). 

7 Circular, Shawal 21, 1248 (F.O. 78-231). 

8 Circular to mudirs, Zilhaj 14, 1248 (Abdine Archives). 



inflicted instead . 1 Here is something real to justify 
Palmerston’s aversion to the pasha’s rule and to rein- 
force political expediency by humane considerations. 

Nor in this matter can the blame fairly be laid upon 
the Sultan’s obstinate hostility, for recruitment might 
have been fairer, less a matter of official greed and 
oppression. In this matter the pasha was certainly 
carried away by his political dreams into a policy which 
reminds one that he was a Turk first and a benevolent 
despot afterwards. Yet, if the cruelty of his methods 
can be put aside, the purpose and the result of his 
military levies were far from wholly evil. Nothing else 
could have done so much to raise the spirit of men who 
had been serfs from before the time when the first of the 
pyramids was built. No one since the Arab conquest 
had ever thought of asking them to fight. They had 
been good to till the fields, to carry burdens, to be 
beaten, to obey commands, to beget children who 
would bear the same painful inheritance. Their terrors 
at being seized and dragged away to serve in the pasha’s 
army was great enough to induce them to hack off a 
finger, to wrench out their teeth, to blind an eye. But 
their reluctance to play the part of men does not 
condemn the pasha for compelling them to play it. 
Nor was this all. Observers agree that the New Model — 
the Nizam jadid — was incomparably less oppressive to 
the general population than the undisciplined foreign 
mercenaries had been. They did not leave a trail of 
havoc behind them in their march. They did not move 
through an Egyptian province as if they were in an 
enemy’s country. The pasha’s military organisation was 
not only an act of power executed in complete disregard 
of the subject’s wishes, but also a measure of education 
and an administrative reform. 

1 To the Minister of Marine, Rabi-al-awal 3, 1251 (Abdine 



Justice, though urgently needing reform, offered 
problems which could not be solved by swift violence. 
It was too closely intertwined with the sacred law for 
the pasha to touch it with anything but great caution. 
In all matters of ecclesiastical law, marriage, divorce 
and, above all, inheritance, the only competent 
authority was the mufti, whose annual appointment by 
the Porte was one of the very few remaining relics of 
Turkish dominion. This official always bought his 
office from the Divan, and so could not be expected to 
be over-scrupulous in his administration of justice or in 
the selection of the subordinate qazis who held posts 
under him. Muhammad ’Ali had the poorest opinion 
of their integrity. To a family quarrelling over the 
division of an inheritance he warmly recommended a 
friendly agreement, for, if they fell into the qazi’s 
clutches, not one alone, but all the claimants would get 
the smaller portion . 1 But although the pasha could not 
interfere directly with their jurisdiction, he did what he 
could to limit its effects. At Cairo and Alexandria he 
established two new courts, entirely freed from the 
trammels of Islamic law, consisting of merchants in- 
stead of theologians, and intended to determine com- 
mercial disputes, especially those arising between 
Muslims and Christians. What was specially note- 
worthy was that these courts did not include even a 
majority of Muslims. At Alexandria, for instance, it 
was composed of nine members, of whom only four 
were Arabs, the others being a Frenchman, a Jew, two 
Levantine Christians and a Greek . 2 

Criminal justice was normally administered by the 
executive authorities. From the first establishment of 
his power, the pasha had done his utmost to repress all 

1 Puckler-Muskau, op. cit. 1, 286-7. 

2 St John, op. cit. u, 429. Campbell’s Report on Syria, enc. in 
despatch, August 23, 1836 (F.O. 78-283). 


crimes of violence. Missett in 1813 had commented on 
the remarkable fact that the inhabitants of Cairo for the 
first time for many years enjoyed complete personal 
security. 1 This was only accomplished by the exercise 
of much severity and many executions. The Bab-uz- 
zuwaila, where public hangings took place, was at one 
time constantly adorned by the corpses of malefactors. 
The pasha’s sentences were arbitrary, and he would at 
times give decisions hardly reconcilable with the ideas 
of Europe. A man convicted of theft from the musket 
factory, for instance, was to work for life in irons if 
young, but if old was to be hanged as a salutary 
example. 2 But there is little to suggest that the ad- 
ministration of criminal justice was more ferocious than 
it had been in England at all events in theory until 
Peel’s reforms, and as time went on it became notice- 
ably milder. Labour in chains was frequently sub- 
stituted for the rope, and the executioner at Cairo told 
Bowring that he had little then to do. 3 

Slavery and the slave trade were far too well- 
established institutions to be abolished by Muhammad 
’Ali, even had he desired to do so. He had been 
familiar with them from childhood; they represented 
the immemorial practice of the East; they violated no 
moral sentiment of oriental minds ; until recently they 
had not seriously disturbed the much more squeamish 
conscience of the West. Nothing then was done, or was 
likely to be done, to limit the slave market of Cairo or 
reduce the power which Muslim law conferred upon 
the master over his slave. In 1836 the Russian consul- 
general, Du Hamel, raised the question, and asked the 
pasha whether he could not take away from masters 
the power of inflicting death and wanton ill-treatment. 

1 Missett, November 9, 1813 (F.O. 24-4). 

2 To Habib Effendi, Zilhaj 26, 1252 (Abdine Archives). 

3 Bowring, Report (Pari. Papers, 1840, xxi, 123). 



Muhammad ’Ali thought that something might perhaps 
be done in the case of male slaves, but gave no hopes of 
his being able to interfere with the treatment of women 
slaves. The harims, he said, were sacred places to which 
no stranger could penetrate. 1 There the matter rested. 
The conflict of external policy which speedily arose 
between him and the European powers must have 
strongly indisposed him to touch a thorny question in 
which he had no personal sympathy for reform. 

The slave trade itself was one of the oldest branches 
of traffic in the countries which he ruled. Slave hunts 
were periodically conducted in the Sudan and the 
country to the southward, and the captives carried up 
in caravans of considerable size. Precise information 
on this subject was very hard to get, but a French 
enquirer at the time of the French occupation gathered 
from the Copt writer who had registered the slaves 
brought up to Cairo for thirty years that they had 
hardly ever exceeded 4000 in a year. 2 3 On the establish- 
ment of Muhammad ’Ali’s power in the Sudan, it is 
likely that for a time the number considerably in- 
creased. The reader may remember that the pasha had 
cherished the design of forming a great army of them. 8 
Every autumn raids were carried out, and for a time 
the slave trade in that region was made a government 
monopoly. 4 * Muhammad ’Ali’s southern conquests 
were not the only cause making for the expansion of 
this traffic. The Russian occupation of Georgia and 
Circassia greatly reduced the supply of slaves sent from 
those regions to Constantinople, and so increased the 
demand for such slaves as could be supplied from 

1 Campbell, December 24, 1836 (F.O. 78-284). 

2 Frank, Commerce des Negres au Kaire, pp. 19 sqq. 

3 Vide p. 63, supra. 

4 To the sar’-askar, Kordofan, Rabi-al-awal 15, 1237 (Abdine 



Egypt. Then too the matter became much better 
known. The regularity of the new government per- 
mitted Europeans to travel through the Sudan with 
safety. A certain Dr Holroyd, for example, furnished 
details of the slave hunts and of the treatment of the 
captured, which perhaps did not exaggerate the horrors 
of the practice, but which was certainly employed by 
Palmerston to prejudice the English public against 
Muhammad ’Ali’s government. 1 After the pasha’s 
prolonged tour in the Sudan in 1838, he took measures 
to limit the slave trade there. The land revenue, which 
till then had been paid in slaves, was ordered to be 
collected in future in grain and other produce; but, 
although this no doubt did some good, the barbarous, 
time-honoured custom of slave raids long continued to 
persist. 2 

A strong contrast to the pasha’s attitude to slavery 
and the slave trade is afforded by his policy in matters 
of sanitation and education. He was not much im- 
pressed by the western arguments in favour of universal * 
liberty. But he was quite sure that western physicians 
knew more than Arab hakims , and that his people 
would greatly benefit from western knowledge. In 
sanitation and medical organisation he placed great 
and deserved faith in the French doctor, Clot Bey, 
under whom was established a medical school at Abu- 
zabel. The general level of education rendered this 
attempt premature. The chief instructors were French- 
men who knew no Arabic; the students were Arabs 
who knew no French. The net result can hardly have 
exceeded the production of a number of “surgeons” 
with no adequate knowledge of western medicine. It 

1 Bowring, Report {Pari. Papers, 1840, xxi, 83 sqq.). Campbell 
to Bidwell, December 1, 1837 (F.O. 78-322). 

a Campbell, March 15, 1839, and Barnett, April 19, 1843 
(F.O. 78-373, 541). 



would have been wiser no doubt to have begun by 
sending a limited number of men to be trained abroad. 
But the pasha wished to have surgeons and assistant- 
surgeons attached to the various units of his army and 
insisted on immediate provision for them. He was 
however always eager to encourage skilled visitors. In 
1836 an oculist, Dr Charles Nayler, visited Alexandria. 
He was so successful in treating cases of ophthalmia 
that he was besieged by men of all classes, seventy or 
eighty daily gathering outside his house in the hopes of 
attracting his attention and benefiting by what seemed 
to them more than human skill. The pasha, in the hope 
of securing his services, offered him a salary of ■£ r 200 a 
year. 1 

One of the most horrible sights of Cairo had been 
“the hospital of the Moristan” — a charity attached to 
one of the mosques, where one could visit sick men 
stinking with filth and over-run with vermin, or, better 
still, peer through square gratings at chained and naked 
maniacs. These most pitiable of beings were in charge 
of an old Arab who showed them off in the hope that 
visitors would make him presents. At Clot Bey’s sug- 
gestion the pasha agreed to the abolition of this medieval 
relic and ordered another hospital to be prepared in 
the great square of Usbekiah. 2 

Another illustration of Muhammad ’Ali’s readiness 
to adopt improvements is afforded by the Board of 
Sanitation. In 1831 a very severe epidemic of cholera 
broke out. It had been brought by pilgrims from the 
Hijaz to Suez, where 150 died in two days, and then a 
fortnight later it broke out suddenly at Cairo. In the 
hope of preventing the disease from spreading to 
Alexandria, the pasha invited the aid of the consuls- 

1 Campbell, October 5, 1836 (F.O. 78-284). 

8 St John, op. cit. u, 309. Bowring, Report (Pari. Papers, 1840, 
xxi, 141). 


general, placing all his troops in the neighbourhood at 
their disposal and giving them complete liberty in the 
matter of expenditure. They accepted the call, although 
they seem to have despaired of staying the disease. Two 
cordons were established between Cairo and Alexan- 
dria. But, as might no doubt have been expected, the 
cordons themselves speedily were attacked. In less 
than a week 800 soldiers were in hospital, the physicians 
and apothecaries either died or ran away, all public 
services fell into confusion, and protective measures 
were abandoned. Before the epidemic ceased, not far 
from 9000 persons had perished at Cairo, and over 
1500 at Alexandria. At this time the populations of 
the two cities were computed at 300,000 and 90,000 
respectively. 1 

Cholera was only an occasional scourge on so large a 
scale as this, and did not again become epidemic until 
1 848.2 But bubonic plague was a constant source of 
terror. Readers of Eothen will remember how the 
Franks were accustomed, when plague was abroad, to 
shut themselves up in strict seclusion, while the Muslims 
gloomily attempted to ignore the danger that encom- 
passed them. But it was not Muslims of all ranks who 
in such times were to be met striding defiantly through 
the streets. Such confidence in the irrevocable decrees 
of Allah was displayed by few save the poorer sort, 
whom narrow circumstances made less careful of the 
continuance of their earthly penury. Those whom 
wealth allowed a foretaste of the delights of heaven 
were as cautious as the unbelieving Franks themselves. 
The pasha could be visited by none — not even the 
consuls-general were allowed to break his quarantine. 
The public offices were closed. Business came to a 

1 Barker, August 1 8 and 23 and September 2 ; Barker to Gordon, 
September 29, 1831 (F.O. 78-202). 

2 Murray, July 26, 1848 (F.O. 78-757). 


stand. 1 The worst outbreak was the one that fell upon 
Lower Egypt in 1835. It was thought even more 
destructive than the Great Plague that had raged forty 
years earlier. In three months the casualties were 
announced as 31,000 at Cairo alone. But Campbell 
believed that they had been much more numerous. 
He reckoned that more than this number of Muslims 
alone had died. In a single great household 135 deaths 
had occurred. Twelve hundred Muslim houses were 
shut up because every inhabitant was dead. A quarter 
of the Copts perished, adding 20,000 to the roll. 2 

Since quarantine was the only protective measure 
that had as yet been adopted against the plague, the 
pasha had again invoked the services of the consuls- 
general, without whose aid and concurrence it would 
be difficult and perhaps dangerous to enforce quaran- 
tine on a large number of European ships and sailors. 
The consuls had therefore formed a committee which 
was known at different times as the Board of Sanitation 
and the Committee of Health. A lazaretto had been 
built, near the place where Cleopatra’s Needle then 
stood, on the shore of the New (or eastern) Port at 
Alexandria, where all vessels undergoing quarantine 
were required to anchor; 3 and the chief of police at 
Alexandria was warned to enforce all the rules that the 
consuls might adopt. This was no easy matter. The in- 
habitants were unwilling to follow rules, the object of 
which they could not understand, and which they 
fancied were contrary to the rules of their faith. The 
avoidance of contagion, the pasha declared, is not con- 
trary to the law; he promised to obtain a fatwa or 
declaration to that effect from the theologians. The 

1 Salt, June 15, 1816 (F.O. 24-6). Campbell, March 29 and 
April 15, 1835 (F.O. 78-257). 

2 Campbell, June 25, 1835 (F.O. 78-257). 

3 The same, October 16, 1835 (F.O. 78-260). 


citizens, he concluded, “are like beasts, unable to dis- 
cern good and evil”. 1 A little later another committee 
was formed, with Campbell at its head, to improve the 
general sanitation of Alexandria. A large number of 
“filthy Arab huts” were destroyed; the old ruined 
ditch, full of stagnant water, was filled up ; the govern- 
ment tannery standing in the middle of the city was 
removed; a broad road was cut from the European 
quarter to the Custom House. 2 In 1837, despite the 
constant influx of pilgrims from plague-stricken regions, 
Campbell was able to report that plague had dis- 
appeared. He ascribed it, of course, to the system of 
quarantine, which had been strictly maintained. “The 
pasha leaves the whole affair”, he writes, “in the hands 
of the consular Board of Health, and not only puts in 
force . . . every measure ordered by the Board, but he 
moreover supplies without ever making any difficulty 
all the funds required ... for the service of the lazzaretto, 
and which is very costly from the great number of 
Europeans employed in that service.” 3 

The recall of Campbell and the events of 1839-40 
broke up this organisation. Campbell’s successor, 
Hodges, was much more interested in obtaining infor- 
mation about the defences of Alexandria than in aiding 
the administration. A new Board of Health was there- 
fore set up, to which Muhammad ’Ali nominated three 
of the consuls-general, but over which the consuls- 
general as such had no control. Disputes followed about 
the proper composition of the board. Physicians too 
began to question the efficacy of quarantine and to 
suspect that plague might be conveyed by other means 
than personal contact. The old system was therefore 
relaxed and finally abandoned. But it exhibits another 

1 To the head of the Divan, Shawal 1 3, 1250 (Abdine Archives). 

2 Campbell, October 16, 1835 (F.O. 78-260). 

3 The same, November 7, 1837 (F.O. 78-321). 



undeniable proof of the pasha’s willingness to adopt 
European ideas and guidance where he believed they 
might be of real service. 

But his educational measures afford the most re- 
markable evidence of his policy of reform. Cairo was of 
course one of the great centres of Islamic culture. To its 
great university housed in the old A 1 Azhar mosque 
came students of all the nations of Islam. But its 
organisation and studies were alike medieval. It bred 
theologians and ecclesiastical lawyers. It did not breed 
men of affairs and administrators. All western know- 
ledge was completely excluded. New schools were 
needed to provide the wider outlook which the pasha’s 
administrative ideas required. While Muhammad ’Ali 
continued to protect and maintain the ancient uni- 
versity, he set up beside it a whole series of institutions 
designed indirectly to modify and modernise the 
popular mind. An English contemporary well summed 
up his purpose and attitude. Whereas Sultan Mahmud, 
he says, by his sudden, violent reforms had weakened 
'the allegiance of the Turks, Muhammad ’Ali had 
always maintained a high character among Muslims 
“ by adopting the only wise course to be pursued with a 
nation in so low a scale of civilisation. By a system of 
gradual ameliorations, without violating religious pre- 
judices, he has laid the foundation of a permanent 
reform in the institutions of the nation, trusting to the 
progress of knowledge from the general establishment 
of public schools throughout his government for the 
final establishment of his plans of reform”. 1 

This policy seems to date from about 1 820 and was in 
origin the natural corollary of the reform of the army. 
The introduction of European methods of organisa- 
tion and training clearly demanded officers capable 
of studying European military science, engineering, 

1 Memorandum, Thurburn, October 14, 1836 (F.O. 78-295). 


mathematics.The first indication that this was recognised 
seems to be the establishment of an Italian, Costi by 
name, in the citadel of Cairo, to teach drawing and 
mathematics. Then come orders for the teaching of 
Italian — the lingua franca of the Levant — and demands 
for teachers of French and Turkish, and a capable 
engineer. 1 From this simple beginning arose schools 
designed to train officers for the five branches of the 
pasha’s service — the artillery, engineers, cavalry, in- 
fantry, and marine — under European direction. 

In order to broaden the basis of instruction, a con- 
siderable number of young Egyptians were sent to 
France, and a few to England, to be educated at the 
pasha’s expense. The fruits of this were seen in 1833 
when a Polytechnic School was established as a train- 
ing school for the officers’ colleges. The teaching staff 
included two Europeans only, one for chemistry and 
one' for mathematics. Beside them were four Arme- 
nians, one of whom had spent ten years at Stoneyhurst, 
and six Muslims, three of whom had been educated at 
Paris and three in England. 2 This expansion was 
followed up by the establishment of several primary 
schools in each mudirliq, with two large “preparatory” 
schools, one at Cairo and one at Alexandria, designed 
to feed the Polytechnic School. Admission to the 
schools was in effect reception into the pasha’s service. 
The pupils were fed, clothed, and lodged at the public 
charge, and received besides small monthly allowances, 
rising in amount as the boys passed from class to class. 
Their future career, the branch of service into which they 
were drafted, the technical training which each received, 
was a matter for the determination of the pasha or his 
officials. Egypt was the first oriental country in which 

1 Letters of Zilhaj 4 and 8, 1235, and Rabi-al-awal 5, 1236 
(Abdine Archives). 

* Campbell, November 14, 1833 (F.O. 78-228). 



anything like a regular system of westernised education 
was established. 1 Bowring was right in criticising the 
scheme as founded on too narrow a basis of primary 
education, and as designed to secure superior education 
for the few instead of providing a universal system for 
the many. But the pasha cannot reasonably be con- 
demned for not adopting a system which the most 
advanced western nations had not yet adopted. 

This foundation of schools and colleges was accom- 
panied by the establishment of printing presses and the 
appearance of a newspaper and a gazette. By the end 
of 1837 seventy-three oriental works had been printed 
at the press set up at Bulaq, then the suburb of Cairo 
abutting on the Nile. These included translations of a 
good many technical works for the use of the new 
schools. 2 The pasha projected a newspaper to be printed 
in French and Arabic. 3 The European Press existed at 
Alexandria in 1824, an d in that year printed a de- 
scriptive poem by Salt, the consul-general. 4 At the 
same time the position of the European and the 
Christian were transformed. Before Muhammad ’Ali’s 
rise to power native Christians had been subject to 
many disabilities. They were obliged to distinguish 
themselves by the colour of their dress. They were 
forbidden to ride horses. They were forbidden during 
the month of the great Muslim fast to eat, drink, or 
smoke in the streets by day, so as not to remind the true 
believer of his compulsory abstinence. 6 The Frank in- 
habitants of Cairo and Alexandria lived in separate 
quarters with guarded entrances, and, when they went 
abroad, wore Turkish dress in order to avoid insult. 

1 Puckler-Muskau, op. cit. 1, 125 sqq. Bowring, Report {Pari. 
Papers, 1840, xxi, 125 sqq.). 

a Medem to Nesselrode, January 12/24, 1838. 

3 St John, op. cit. 1, 54. 

4 Egypt : a descriptive poem. 

6 Politis, op. cit. 1, 1 75. Jabarti, op. cit. vi, 95. 


“As is known to every prudent person”, ran the 
Turkish declaration of war on Russia in 1827, “every 
Muslim is by nature the mortal enemy of infidels, and 
every infidel the mortal enemy of Muslims.” But under 
the pasha’s rule the spirit of the government was trans- 
formed, and that of the people sensibly modified. Two 
of Muhammad ’Ali’s relatives, made prisoners by the 
Russians in 1827, returned from captivity in 1829 full 
of the good treatment they and other prisoners of war 
had experienced. 1 In a year when the Nile threatened 
not to rise to its accustomed height, joint supplications 
were offered on its bank not only by the heads of the 
different Muslim sects, but by Jewish rabbis and 
Christian priests as well. 2 Wolff, the mad missionary, 
was suffered readily enough to preach in the streets in 
an Arabic that no one could understand, but, when he 
placarded Cairo with too legible inscriptions, then 
indeed the pasha requested his departure since nothing 
could be done to protect him from a chance attack. 3 
Throughout the crisis of 1 839-40 Englishmen continued 
to dwell in Cairo and Alexandria without insult. 

This policy was of course unpopular with the shaikhs 
of A 1 Azhar. One popular preacher, Shaikh Ibrahim, 
was bitter in his denunciations. The Jews, for instance, 
secured a monopoly of the butcher’s trade at Alexandria, 
and so imperilled the salvation of all true believers there, 
for they would not use the sacred formula, or turn the 
animal’s head towards Mecca, and the hafts of their 
knives were fastened with only three nails instead of 
five. 4 But the shaikh was exiled to Tunis. 

In all these matters, in compelling toleration, in 
promoting health and knowledge, in dispensing strict 

1 Barker to Gordon, May 20, 1829 (F.O. 78-184). 

2 Clot Bey, Aperpu, 1, 317. 

8 Barker, Syria and Egypt, n, 142. 

4 St John, op. cit. 1, 43. 


justice, in reorganising his troops and building up a 
naval force, in defining taxation, in encouraging new 
crops, in watching closely over the conduct of his 
officials, the pasha had to work against the opposition 
of almost all his own people. Most of his most cherished 
projects were neglected or abandoned by his immediate 
successors. Some, such as the formation of a navy, were 
opposed by obstacles too great to be overcome. Most 
were weakened and impaired by his lack of confidence 
in the future and his sense that whatever was done 
must be done by himself and within the limited dura- 
tion of his own life. So that his work must be adjudged 
hasty, premature, unfinished. But in spite of that, in 
spite of the reaction which followed on his disap- 
pearance, it would be quite unfair to regard his work 
as wasted. The strong impulse which he had given, the 
contacts which he had established with the West, 
continued, so that when at a later time Egypt began 
once more to go forward, she began her new movement 
far in advance of the point at which the great pasha had 
found her, thanks above everything to the cultural in- 
fluences to which he had opened his country so widely 
and so wisely. 

D M 


Chapter VIII 


In the course of the Greek war, both Cyprus and Crete 
had been placed by the Sultan under the care of 
Muhammad ’Ali, presumably because he alone could 
secure them from Greek attacks. In 1830 the govern- 
ment of Crete was formally confided to him. He had 
made it a condition of accepting the charge that he 
should be allowed to send away the Turkish troops 
there and garrison the island solely with his Arab 
regiments. 1 He first sent as commandant a Turk 
named Usman Bey, whom he had sent to Italy and 
France to be educated. 2 The population was mixed, 
rather more than half being Greek by race. Fifty years 
earlier Savary had reckoned the total at 380,000 
persons ; but war, plague and misery had reduced it to 
less than 100,000 at the time when the pasha under- 
took its government. The mixture of peoples promised 
much trouble. Nor was the administration likely to be a 
gainful affair. The revenues were under four million 
piastres against an expenditure of over eleven. It 
seems probable that the offer was accepted because it 
gave the pasha a naval station well to the northward of 
Alexandria. He had been twice warned by Great 
Britain that any violent oppression of the Christian 
inhabitants might provoke the interference of the great 
powers. 3 

1 Barker to Sir P. Malcolm, August 31, 1830 (F.O. 78-192). 

4 Same to same, September 17, 1830 ( ibid .). 

3 To Barker, October 15 and December 31, 1828 (F.O. 78- 



Such a warning was probably quite needless. The 
ruler who protected Christian minorities in Egypt and 
Syria was not likely to persecute a majority in Crete. 
His first act after the proclamation of his farman of 
appointment was to issue an address to the Candiotes. 
He assured them that they had nothing to fear, that 
he would smite any who smote them, and that he would 
establish two councils, one at Canea, the other at 
Candia, composed of both Christian and Muslim 
members, to hear and determine all but strictly legal 
questions such as inheritance. He intended to introduce 
material improvements too, such as a breakwater for 
the harbour of Canea, the re-afforestation of the hills, 
and the extension of cultivation. 1 Another project was 
to develop the port of Suda, both as an entrepot for the 
Syrian trade and as a base for the Egyptian fleet. 2 

In 1833 the pasha visited Crete in person. Campbell 
at his invitation accompanied him and sent to England 
some interesting notes on the island. Under the Sultan 
it had been misruled by three pashas, all frequently 
changed and invariably oppressive. At the time of the 
Greek war the Turks would certainly have been 
driven out but for the troops sent by Muhammad ’Ali. 
Since the island had been transferred to him, he had 
sent as governor Mustafa Pasha, feared by the Turks 
but much esteemed by the Greek inhabitants. He had 
set up the promised mixed councils and two tribunals 
of first instance — one at Sphakia composed entirely of 
Greeks, as no Turks were living there. Loans of money 
and cattle had been made to enable the impoverished 
peasants to recultivate their farms, and a proclamation 
was issued that those Greeks who had fled might return 
and reoccupy their farms on condition of paying out 
those who had bought them at the same price as the 

1 Barker to Gordon, September 8, 1830 and enc. (F.O. 78-192). 

8 Campbell, May 26, 1833 (F.O. 78-227). 


latter had paid. Many had done so and settled down as 
rayas under the pasha’s government. 1 

However, in spite of the mildness of the administra- 
tion, a good deal of discontent appeared. Many refu- 
gees refused to return except with Greek passports as 
Greek subjects, and some landed secretly in the hope of 
raising new troubles. The Candiote emigrants pub- 
lished a journal of their own called the Minerva , the 
main object of which was to arouse discontent in the 
island. 2 The pasha wisely refused to allow the refugees 
to re-enter the island except as rayas, declaring that 
any other course would greatly discontent the 60,000 
Christians who had remained there and not demanded 
a new status. 3 But while the former irregular exactions 
were suppressed, regular taxation was increased. The 
karach or capitation-tax, levied on all Christian subjects 
of the Porte, 4 was collected with greater rigour and 
fewer exemptions. A duty was laid on wine, whether 
made for sale or for the private consumption of the 
maker. The monopoly of selling tobacco, wine and 
leather in the towns was farmed out. Much excite- 
ment arose. Miracles were reported to have occurred 
at different convents, and the peasants began to 
assemble in crowds. All this seems to have resulted 
from the imigrt. propaganda, and, when the time was 
judged ripe for an explosion, an unfortunate Turkish 
traveller was set upon and murdered. The assassin, a 
returned emigrant, was duly hanged. But this was 
the only execution. The remainder of the returned 
emigrants were either deported or permitted to re- 
sume their raya status on condition of their villages’ 

1 Campbell, August 29, 1833 (F.O. 78-228). 

2 The same, August 21, 1833 (ibid.). 

3 The same, August 28, 1833 (ibid.). 

4 The word is properly khiraj, land revenue; but the Turks 
seem to have dropped this use of the word, and applied it to what 
was called the jaziya elsewhere. 



standing surety for their good conduct , 1 and the pasha, 
having issued orders which were intended to increase 
the cultivation of the island, returned to Alexandria. 

But unluckily his orders provided the occasion for 
fresh trouble. One of them directed the appointment 
in each district of two persons acquainted with the laws 
of Egypt, to visit each village, to consult with the rich 
as to the best means of assisting the poor, and to concert 
methods of transferring labour from the more populous 
villages to those in which men were few and lands un- 
cultivated. Although this measure was accompanied 
by others undeniably calculated to benefit the people 
at large, such as the establishment of schools and the 
payment of allowances to the pupils, the Candiotes 
felt that the pasha was seeking to introduce the same 
control of the soil as prevailed in Egypt. They broke 
out into revolt, despite the fact that “the system of 
Mehemet Ali in Candia, however mistaken it may 
perhaps be in some points, breathes a spirit of benevo- 
lence, a liberality of principle in religion, a love of 
justice and an evident desire for the happiness and 
welfare of the people which reflect great credit to 
him ”. 2 

The pasha was indignant at this reception of his 
plans for the betterment of the island and determined 
to make an example of those mainly concerned; he 
ordered a certain number to be hanged if found guilty 
of rebellion, and he told Campbell that he expected a 
certain number of Turks would be found concerned in 
the revolt as well as the Greeks. If so, they too should 
be hanged . 3 In the end thirty-one persons, including 
five Turks, were executed. The French consul, who 
chanced to be an ardent Phil-hellene, said they were 

1 Campbell, August 29, 1833 (F.O. 78-228). 

2 The same, October 10, 1833 (ibid.). 

3 Idem. 


put to death without trial. 1 The pasha seems to have 
been convinced that the whole affair was the work of a 
small number of agitators, and to have determined to 
inflict a punishment which all the protests of all the 
powers could not revoke, so restricted as to give none 
an excuse for intervention, but so extended as to give 
the Candiotes a sharp lesson. If so, he reckoned justly, 
for he had no further trouble in Crete. 

The administration of Crete had been entrusted to 
Mustafa Pasha, who continued to exercise it during 
practically the whole of Muhammad ’Ali’s possession 
of the island. The consuls of the English, French and 
Russian governments all agree that his management 
was mild, popular, and in a high degree successful. 
Of course he was not able wholly to quench political 
discontent. Crete was still a part of Grecia Inidenta. 
There were always societies in Greece eager to add the 
island to their national kingdom. There were always in 
the island a number of persons who dreamt of reunion 
with their fellow-countrymen or the establishment of , 
some sort of independence. 2 But so long as Mustafa 
Pasha ruled Crete, it remained quiet and contented. 
The Russian consul reports that taxes were paid with- 
out resistance, that complete tranquillity reigned, that 
the municipal councils were always ready to do as the 
governor wished. 3 In 1838 he was ordered to Syria to 
take command of troops sent to suppress an insurrec- 
tion that had broken out there. The English consul 
remarks that his departure occasioned “the most spon- 
taneous, disinterested and unequivocal marks of. . . 
affection” from every class of the people. When he left 
Canea, he was accompanied to his boat by the whole 
population, young and old, lamenting with tears his 

1 Campbell, December 31, 1833 (F.O. 78-228). 

2 Lyons, August 2, 1838 (F.O. 32-78). 

3 Thoron to Medem, January io, 1838. 



departure and entreating him to return to them. 1 He 
had in fact protected the Greeks and soothed the 

In accordance however with his custom of believing 
the worst about Muhammad ’Ali’s government and 
designs, Palmerston could not leave the island alone. 
The hanging of twenty-six Greeks and five Turks is 
mentioned by him “if report speaks true” as great 
severity and numerous executions. He suggested that 
the pasha might resign the island to the beneficent rule 
of the Sultan, who might be induced to bestow on it a 
constitution such as that existing in Samos. 2 Several 
discussions followed between Campbell on the one 
hand and the pasha and his chief minister Boghoz on 
the other, but the latter declined absolutely the pro- 
posals made to them. Crete, they said with truth, was 
not like Samos, populated only by the Greeks. There 
lived a large body of Muslims who could not reasonably 
be subjected to Greek rule, while the settlement of 
Greeks in the island showed conclusively that the 
pasha’s rule was neither severe, intolerant nor unjust. 3 
So matters remained till 1840 when the pasha lost 
Crete as well as Syria. Palmerston at once revived his 
former project of establishing the Samian constitution 
there — a plan to which he seems to have been specially 
attached. Perhaps he thought that when Campbell had 
discussed the question with Muhammad ’Ali, it could 
not have been fairly presented; Campbell must have 
been tactless and perhaps unconvincing; and the pasha 
had of course been unwilling to make any real reform. 
But Ponsonby in fact could do no better with the Porte 

1 Campbell, April 24, 1838 (F.O. 78-342). 

2 To Campbell, March 3, 1834 (F.O. 78-244). This island was 
wholly peopled by Greeks and governed in the Sultan’s name by a 
Greek chief. 

3 Campbell, May 17, 1838 (F.O. 78-342). 


than Campbell had done with Muhammad ’Ali. At 
Constantinople as at Cairo the Samian plan was 
deemed unsuitable to Crete — and Ponsonby agreed. 
The Turk inhabitants, he said, could not be subjected 
to Greek tyranny, nor the forts occupied by Greek 
garrisons. It would mean constant rebellions, and the 
island would pass to Greece, or France or Russia. So 
Crete was given back to the Sultan, without the con- 
stitution which Palmerston had at one time judged so 
important to its well-being. 

The administration of Crete by Muhammad ’Ali 
deserves to be regarded as a success. But the same can 
scarcely be said of Syria, where the task was heavier, 
more complicated, and worse controlled. In Crete the 
Turks formed a minority, and submitted with what 
grace they could to the reforms which the viceroy 
ordered and Mustafa Pasha carried into effect. In Syria 
the population was still more divided by race and creed. 
It was intensely fanatical. Both main divisions were 
cloven by the teaching of rival sects. Out of a total of 
1,800,000, nearly a million were followers of the 
Prophet. The 600,000 Christians were divided between 
the Catholic and the Orthodox Churches; the re- 
mainder consisted of minor groups. So that the Muslim 
majority could lord it over these small and disunited 
bodies with little fear of possible reprisals. Secular 
authority was disputed and defied by territorial chiefs 
who had ruled with little interference from the Turkish 
authorities. Above all, while Crete was sheltered from 
external interference by the pasha’s fleet, Syria lay 
open to the active intrigue of the Sultan’s agents and 
the attacks of the Sultan’s forces all along a lengthy 
land frontier. 

The Turks had divided the region into four pasha- 
liqs — Acre, Tripoli, Damascus and Aleppo. But Ab- 
dullah, the gallant defender of Acre against Ibrahim, 


had obtained, besides Acre, Tripoli and the districts of 
Nablus and Jerusalem. But within his government 
were a number of chiefs who paid no obedience to his 
authority. One was the Amir Beshir, the prince of 
Lebanon. Another was the Bedouin chief Abu Ghosh, 
established between Gaza and Jerusalem. A third was 
Mustafa Barbar who had driven the Turkish deputy 
out of Tripoli and whom Abdullah had to expel by 
force of arms. In the hills of Nablus was a shaikh 
whose authority had been recognised by the Porte 
itself. Aleppo was torn in pieces by rival Muslim 
factions. In Damascus Christians and Jews lived in 
perpetual fear for their lives and goods. Not even the 
pasha himself was safe. “The Porte on learning that the 
citizens of Damascus had cut his pasha’s throat, merely 
sent them another .” 1 The reader of Eothen will re- 
member with what delight a raya saw Kinglake 
boldly walking along the raised footway which the 
true believers had been wont to reserve for themselves. 
Indeed, as the French agent Boislecomte most judi- 
ciously observed, religion in Syria occupied much the 
same position as race in Egypt. In the latter a man 
belonged to the ruling or the subject class according as 
he was born a Turk or an Arab. In Syria Turk and 
Arab stood on a common footing, and a man’s degree 
depended on his being bred a Christian or a Muslim . 2 
But this solidarity of feeling within the two faiths did 
not extend to the desert Arabs on the east and south. 
In the days of Wahabi greatness ibn Saud in two days 
and a half over-ran at least 1 20 miles and plundered 
thirty villages , 3 while the Bedouin were always raiding 
the settled country and driving the inhabitants west- 
wards, so that the eastern borderland was full of 

1 Douin, Mission de Boislecomte, pp. 220 sqq. 

2 Idem, p. 199. 

8 Burckhardt, Nubia, p. xli. 


deserted villages. Nor had the Turkish government 
been in any way calculated to reform these evils. 
“Everybody knows”, wrote Campbell with unhappy 
truth, “that the few pachalics of which the Porte could 
dispose . . . were put up to the highest bidder at Con- 
stantinople every year, who had no other object in view 
but that of making a fortune upon the poor and un- 
fortunate population which thus became a prey to 
their rapacity and avarice. . . . Hence their utter negli- 
gence in every part of the administration, their total 
indifference to the shameful and numerous depredations 
of the richer over the poorer classes, and the total 
absence of troops and means (for purposes of private 
gain) to repel the attacks of the surrounding Arabs .” 1 

Stern military rule was the first condition of restoring 
peace and order to this unfortunate country, and that, 
at all events, was what the inhabitants might rely on 
getting from their new ruler. Ibrahim was entrusted 
with the government of the territory which he had 
conquered. The governor of Damascus, Sharif Bey, 
coolly reckoned that the tranquillity of the city would 
not cost the citizens more than one head a month . 2 
Every revolt that occurred was made the occasion of 
disarming the guilty district. Every inroad of Bedouins 
led to swift reprisals. The Syrians too were not merely 
to be protected from themselves or their neighbours. 
They were to be taught to protect themselves. How- 
ever averse they might be to military service, military 
service would be forced upon them. Under Ibrahim’s 
drill-sergeants they would at least acquire habits of 
discipline to which they had long been strangers. The 
Syrian, like the Egyptian, conscription was cruel, 
capricious and corrupt . 3 It was conducted, as in 

1 Campbell, Report on Syria (F.O. 78-283). 

2 Douin, op. cit. p. 231. 

8 Cf. p. 226, supra. 


Egypt, by a series of haphazard seizures from which the 
wealthier redeemed themselves by bribing the in- 
ferior agents who had seized them. But it must have 
carried with it fortifying virtues in which the Syrian 
was peculiarly lacking. It was natural for Palmerston 
to condemn on moral grounds what he had political 
reason for disliking. But after the passage of a century 
the need for moral indignation has vanished, and later 
experience suggests that the Syrian would have lost 
little had he enjoyed a longer period of Ibrahim’s 

As in Egypt religious toleration was enforced in a way 
till then completely unknown. Once the ulemas and 
theologians of Damascus waited on Ibrahim to com- 
plain that Christians were suffered to ride on horse- 
back and that the due distinction between Muslim and 
infidel was being obliterated. Ibrahim ironically 
agreed that a distinction should be kept, and proposed 
that in future the Muslims should ride dromedaries, 
which would set them high above all Christians. 1 On 
one tragic occasion, commemorated by Robert Curzon, 
Ibrahim himself attended the Miracle of the Holy Fire 
at Jerusalem. 2 These two measures — conscription and 
toleration — incensed the whole Muslim population 
against the new government. 3 When Marmont visited 
Syria in 1834 he found all the Turks very hostile to 
Ibrahim, although in the Ottoman provinces through 
which he had just passed the Turks there were quite as 
bitter against the Sultan. 4 The English consul at 
Aleppo describes Syrian feelings as those of disgust and 
even hatred. 5 

1 Campbell, March 17, 1834 (F.O. 78-245). 

2 Curzon, Monasteries in the Levant, chapter xvi. 

3 Cf. p. 1 56, supra. 

* Campbell, October 9, 1834 (F.O. 78-247). 

8 Picciotto to Campbell, March 3, 1835 (F.O. 78-257). 


This feeling was no doubt strengthened by another 
most disconcerting innovation — the discouragement of 
bribery in matters of justice. All the English consuls in 
1836, who assuredly cannot be cited as witnesses 
favourable to Ibrahim’s administration, are agreed on 
this point. The least favourable admits that it was 
diminished. Another believes that it still existed but in 
a very limited and secret way. A third declares that it 
was rigorously checked. 1 All agree, though reluctantly, 
that justice was no longer an ideal applicable to Mus- 
lims only. One laments the absence of a code of laws. 
But even he recognises the establishment in the larger 
towns of courts, like the new courts established in 
Egypt, in which Jews and Christians were admitted to 
sit among the judges. The new system had indeed the 
merit of a wide elasticity. The plaintiff had the option 
of laying his grievances before either the old mufti’s 
court or the chief executive official. If he selected 
the former, the sentence had to be reported to the 
executive government which could confirm or refuse 
to execute the decision. If a plaintiff applied to the 
executive official, the latter could, if the case were 
simple, hear and decide it himself; if it demanded a 
knowledge of Islamic law, he could refer it to the 
mufti’s court; if it was a case of complicated accounts or 
commercial usage, he could refer it to the new courts. 
The system of justice contained thus a new and most 
important element. It provided that there should be a 
much greater probability that a non-Muslim party 
should receive an impartial consideration of his case. 
It is perhaps worth noting that under the older system 
which had been displaced the evidence of a non- 
Muslim was completely inadmissible against a true 
believer. 2 

1 Answer to Query 1 1. Campbell, July 31, 1836 (F.O. 78-283). 

* Query 10 (ibid.). 


One of the Syrian consuls summed up the results of 
the establishment of Muhammad Ali’s government as 
including security from arbitrary acts — except in the 
case of the conscription — security of property, a new 
liberty of religion, of life and of amusements, a fair 
distribution of taxes, and in general as near an ap- 
proach to the liberty enjoyed under a free government as 
could be attempted. In many respects he found the 
administration to have been improved beyond what 
anyone could have expected. But, he adds, the people 
“do not appreciate it and are prone from their old 
feelings, habits and ideas to turn it solely to their 
individual profit ”. 1 Another observes that “native 
capitalists now venture to embark their fortunes in 
commercial speculations which formerly they did not 
venture to do ”. 2 

Trade and especially agriculture were much increased. 
The rent of land in some areas more than trebled. This 
change was occasioned, we are told, by the rise of 
competition. Round Aleppo, for instance, the rate of 
rent increased because lands were no longer allotted on 
a system of favouritism and influence, and that despite 
the fact that lands abandoned owing to Bedouin raids 
had been brought again under cultivation . 3 En- 
deavours were made to bring the nomads into more 
constant commercial relations with the settled in- 
habitants, to press eastward the line dividing the desert 
from the sown, and to induce the Bedouin themselves 
to take to cultivation. “This system, if steadily pur- 
sued”, Werry writes, “must be productive of great 
benefit to the Syrian and Arab populations thus 
brought into contact for a peaceable object.” In the 

1 Werry, answer to Query 27. Campbell, July 31, 1836 (F.O. 

2 Moore, answer to Query 20 (ibid.). 

3 Werry, answer to Query 9 (ibid.). 


rich and extensive plain of Adana, for instance, peopled 
by a mixture of Anatolians, Turkomans and Kurds, and 
previously given over to anarchy, the pastoral tribes were 
induced to spend part of the year in cultivation. 1 

It is impossible to judge with precision how the taxa- 
tion levied under Ibrahim compared with what had 
been gathered in under the previous rule. The revenue 
brought into the public treasuries certainly rose. 
Collections were more regularly made and more 
closely controlled. At least one new tax, the firdeh , was 
established. This was a capitation tax which (like the 
income tax of Great Britain) had originally served to 
provide extraordinary revenue in time of war. Muham- 
mad ’Ali made it a regular source of income. At first 
it was based on an estimated payment of 50 piastres a 
head, but this was soon modified by the establishment 
of a sliding scale ranging from 30 to 500 piastres, 
according to the individual’s supposed wealth. On this 
basis an assessment was made on groups of families. 
The group had to provide the amount appointed, but 
were free to apportion it among the members as was 
thought best. Under this arrangement, it is said, the 
poor were often quite exempt while the rich paid more 
than the maximum rate. 2 

The infidel poll tax, in Syria as in Crete miscalled the 
khiraj, was collected under special farmans issued by 
the Porte and remitted to Constantinople for the 
Caliph’s use. This was collected at the rate of 15, 30 and 
60 piastres, according to the wealth of the raya. But 
the officials charged with the collection had always 
made it a pretext for laying the Christians under 
private contribution. In 1835 special measures were 
taken to end this abuse. 3 

1 The same, answer to Query 2 7. Campbell,July 31, 1836 (F.O. 


* Werry, answer to Query 8 (ibid.). 

3 Idem. 


As elsewhere, the miri or land revenue formed the 
main financial resource of the country. But it had been 
governed by no rule, and was based upon no actual 
measurement of the land. The nominal unit employed 
was the extent which in any region a pair of oxen could 
plough in a day — a system which gave the largest 
possible opening to evasion and fraud. No attempt had 
ever been made to fix an assessment on specific land 
holdings, but the governor of a district was required to 
find miri to such and such an amount, which was by 
him distributed among the villages in his jurisdiction, 
and by the villagers among themselves. The basis of 
sound administration — a cadastral survey — was com- 
pletely lacking, though one may reasonably suppose 
that had Muhammad ’Ali’s rule continued, that reform 
would have been extended from Egypt to Syria . 1 

The land revenue administration does not seem to 
have provoked complaints or ill-feeling. But military 
requisitions are frequently mentioned by the consuls as 
excessively vexatious. Provisions were requisitioned at 
less than market price in order to supply marching 
detachments; trees might be cut down for firewood; 
beasts of burden might be seized to provide transport 
for a considerable distance; and although some pay- 
ment was made for them, it never rose to the peasant’s 
demand or compensated the owner for having to follow 
his beast a long way in order to get him back. Allied to 
these was the impressment of artisans to work on the 
fortifications which Ibrahim constructed. The men 
were paid about half the market rate and might be 
detained indefinitely . 2 

For many reasons then Ibrahim’s administration in 
Syria ran far less smoothly and successfully than his 
father’s in Egypt. His conscription alienated all the 

1 Campbell, Report on Syria (F.O. 78-283). 

2 Idem. 


Muslim classes, which alone were exposed to it. Tolera- 
tion enraged every fanatic in the country. His requisi- 
tions annoyed the peasant and the artisan. His severity 
alarmed officials, muftis, qazis, fearful for their time- 
honoured but iniquitous perquisites. Added to this was 
the fact that he was regarded as a foreign ruler, bring- 
ing with him maxims and principles of government 
borrowed from Egypt. The Arabs of Syria had always 
regarded the Arabs of Egypt as immeasurably their 
inferiors. Ibrahim’s conquest enabled the latter to 
return this scorn with interest. The fellah soldiery was 
not nearly so forbearing to the Syrians as he was to his 
own villagers . 1 Ibrahim might establish a system of 
posts between the chief towns, but men distrusted it and 
continued to send their letters by messengers hired 
specially for the purpose . 2 

Another source of difference and difficulty lay in 
Ibrahim’s political ideas. He, far more than his father, 
cherished the idea of reviving the Arab caliphate. 
Muhammad ’Ali was not himself strongly possessed of 
this design, although from time to time he would play, 
with it. He fluctuated between political independence 
and the reformation of the Turlush empire as his great 
political aim. To him the Arab always appeared an 
inferior race, in need of a prolonged and vigorous edu- 
cation. Under him the Arab was never admitted to 
high office either in administration or in the army. His 
son, on the contrary, encouraged the Arab. A French 
observer, Boislecomte, relates that this involved him in 
difficulties of military administration. He loved to live 
freely with his men, sometimes even sporting with them, 
and always praising the race from which they sprang, 
and contrasting it with the stupid, incorrigible Turk. 

1 Douin, op. cit. p. 240. 

2 Werry, answer to Query 12. Campbell, July 31, 1836 (F.O.